ADJUNCT SOCIALIZATION WITH SOCIAL MEDIA: THE MOODLE SOCIALWALL FORMAT By Jessica M. Wicks A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Educational Psychology and Educational Technology Ð Doctor of Philosophy 2015 ABSTRACT ADJUNCT SOCIALIZATION WITH SOCIAL MEDIA: THE MOODLE SOCIALWALL FORMAT By Jessica M. Wicks The growing adjunct faculty presence in higher education highlights a need for research -based onboarding. To explore social media as a low -cost solution for the socialization of adjunct faculty, specifi cally in the form of a new plug -in for Moodle, 77 community college adjunct faculty members took part in a randomized, pre -test post -test control g roup design intervention with the ne w Moodle Socialwall format plug -in. This research was framed by a model of socialization proposed by Schrodt, Cawyer, & Sanders (2003) in which effective organizational socialization is comprised of three essential dime nsions (ownership , adequate information , connectedness ). Though social media has been illustrated to support connection, information needs, and ownership in other contexts, analyses of quantitative data did not find statistically significant effects in this study . A small subset of intervention participants did report that the Socialwall aided their socialization by providing access to relevant information and connection with colleagues. This verbal subset of users also expressed factors related to the socialization of adjunct faculty they felt needed further exploration: adjunctsÕ personal initiative, institutional acknowledgement of adjunctsÕ diverse commitments, access to information and feedback related to adjunctsÕ instructional efficacy, formal in teractions for adjuncts with academic departments, and opportunities for college inclusion that were accessible in the context of adjunctsÕ schedules and commitments. This research provides a cautionary note about the potential of social media tool s for institutional use with adjunct faculty along with notes on the time and attention that must be given to implementation and intervention design to effectively estimate its impacts. Keywords: adjunct faculty, community college, contingent faculty, learning management system, LMS, Moodle, organizational socialization, social media, socialization, Socialwall iv ÒThink for a minute, darling: in fairy tales itÕs always the children who have the fine adventures. Ó - Audrey Niffenegger, The Time TravellerÕs Wife IÕm still flying, Mom. Thank you for the lift. -------------------------- Even after all this time the Sun never says to the Earth, ÒYou owe me.Ó Look what happen s w ith a love like that. It lights the whole sky. - Daniel Ladinsky, The Gift: Poems by Hafiz, the Great Sufi Master Husband, I live to let you shine . v ACKNOWLEDGEMENTS I want to thank the members of my guidance and dissertation committees: Drs. Christine Greenhow, Patrick Dickson, John Dirkx, Cary Ro seth, and Kelly Mix. Your support and assistance has aided my learning and growth. Dr. Greenhow deserves a special thank you as my advisor and chair, acting as both a champion and a critic. She consistently encouraged more effective , connected products from my work . Dr. Robin Dickson also deserves a special thank you for coordinating our inaugural cohort of the EPET hybrid Ph.D. In addition to just getting us through the details, Robin simply cared about (me) us and my (our) progress. That was a kindnes s that will never be forgotten . I want to thank the other members of the 2010 cohort, too. YouÕre wonderful. All of you. Thank you to my family. YouÕve been unyielding in your support. I know you had to give up a lot of me for me to have this. Thank yo u also to my supervisor Anthony Freds and my colleagues Dr. Scott Mertes and Dr. Lori Cortez for the encouragement . Thank you, Eric, for the support. Thank you Dr. Sharma and Emre for the consulting. Thank you to SimonÕs Rock (now Bard College at SimonÕs Rock ) and to the fictional Dr. Marlena Evans. Without you , I would have never pictured myself as a Ph.D. Lastly, I extend gratitude to Mid Michigan Community College and the adjuncts employed there that allowed me to pilo t t hese concepts in practice. I am confident that the data collected through this research will aid us in supporting adjunct faculty and in thinking more purposefully about the role of technology in community development. I truly appreciate each individual that took the time to respond to my inquiries. vi TABLE OF CONTENTS LIST OF TABLES .............................................................................................................. x LIST OF FIGURES ............................................................................................................ xi CHAPTER I ....................................................................................................................... 1 Introduction ....................................................................................................................... 1 Definition of Key Terminology ......................................................................................... 2 Adjunct faculty ........................................................................................................... 2 Social media ................................................................................................................ 2 Socialization ................................................................................................................ 3 Research Goals ................................................................................................................... 4 Problem Statement ............................................................................................................ 5 Statement of Purpose ........................................................................................................ 8 Research Questions and Hypotheses ............................................................................... 8 Overview of methodology .......................................................................................... 9 Rationale and significance ....................................................................................... 10 Role of the researcher .............................................................................................. 11 Researcher assumptions .......................................................................................... 12 Organization of the Dissertation .................................................................................... 14 CHAPTER II ................................................................................................................... 15 Literature Review ............................................................................................................ 15 Organizational Socialization ........................................................................................... 16 Faculty socialization tactics ..................................................................................... 17 Orientations .......................................................................................................... 18 Mentoring .............................................................................................................. 19 Adjunct certification programs ............................................................................ 21 Limitations in socialization research ...................................................................... 23 Conclusions ............................................................................................................... 24 Social Media ..................................................................................................................... 25 Social media in teaching and learning .................................................................... 26 Social media in scholarship and identity development ......................................... 28 Social media in social and civic endeavors ............................................................. 29 Conclusions ............................................................................................................... 30 Justification for This Research ....................................................................................... 31 Conceptual Framework .................................................................................................. 33 Thinking about organizational socialization ......................................................... 34 Connecting organizational socialization to social media ...................................... 36 Social media supports ownership ......................................................................... 37 Social media supports access to information ....................................................... 38 Social media supports connectedness .................................................................. 40 Benefits of the new Moodle Socialwall format plug -in ....................................... 41 vii Conclusion ........................................................................................................................ 43 CHAPTER III .................................................................................................................. 45 Methodology ..................................................................................................................... 45 Rationale for Research Approach .................................................................................. 47 Research Sample .............................................................................................................. 47 Sampling design ........................................................................................................ 48 Power analysis .......................................................................................................... 49 Instrumentation ............................................................................................................... 49 Scale Reliability ............................................................................................................... 51 Procedures ........................................................................................................................ 52 The Moodle Socialwall format plug -in ................................................................... 52 Intervention prompts ............................................................................................... 56 Pilot of survey instruments and prompts ............................................................... 57 Design of survey communication (pre -test and post -test) .................................... 57 Compensation structure .......................................................................................... 58 Recruitment and group assignment ....................................................................... 59 Pre-test and compensation ...................................................................................... 59 Moodle Socialwall intervention ............................................................................... 60 Post -test and compensation ..................................................................................... 61 Data Analysis Methods .................................................................................................... 62 Data analysis for hypotheses ................................................................................... 62 Data analysis for research questions ...................................................................... 64 Reliability and Validity ................................................................................................... 66 Internal validity ........................................................................................................ 66 External validity ....................................................................................................... 68 Qualitative reliability and validity .......................................................................... 69 CHAPTER IV .................................................................................................................. 71 Results ............................................................................................................................... 71 Research Question 1 ........................................................................................................ 71 Hypothesis 1 .............................................................................................................. 72 Hypothesis 2 .............................................................................................................. 73 Hypothesis 3 .............................................................................................................. 74 Comparison of means .............................................................................................. 75 Correlations .............................................................................................................. 76 Summary ................................................................................................................... 77 Research Question 2 ........................................................................................................ 78 Personal characteristics & initiative ....................................................................... 81 Information ............................................................................................................... 83 Connection with colleagues ..................................................................................... 84 Technology ................................................................................................................ 85 Challenges to onboarding ........................................................................................ 86 Summary ................................................................................................................... 87 Research Question 3 ........................................................................................................ 88 Likert -type scores ..................................................................................................... 88 viii Open-ended survey questions .................................................................................. 89 Connection ............................................................................................................ 92 None ...................................................................................................................... 93 Information ........................................................................................................... 93 Lack of participation ............................................................................................ 94 Technology ............................................................................................................ 94 Socialwall comments ................................................................................................ 94 Summary ................................................................................................................... 97 Conclusion ........................................................................................................................ 99 CHAPTER V .................................................................................................................. 103 Discussion and Implications ......................................................................................... 103 Research Question 1 ...................................................................................................... 105 Hypothesis 1 ............................................................................................................ 105 Hypothesis 2 ............................................................................................................ 106 Hypothesis 3 ............................................................................................................ 107 Pearson product -moment correlations ................................................................. 109 Research Question 2 ...................................................................................................... 110 Socialwall usage ...................................................................................................... 110 Anticipated feature use ....................................................................................... 111 Unanticipated feature use .................................................................................. 112 Socialwall interaction themes illustrating socialization ...................................... 114 Ownership ........................................................................................................... 114 Information ......................................................................................................... 117 Connectedness .................................................................................................... 119 Summary ................................................................................................................. 121 Feature use ......................................................................................................... 121 Content of interactions ....................................................................................... 122 Research Question 3 ...................................................................................................... 123 Likert -type data ...................................................................................................... 124 Post -test survey responses and constructs of socialization ................................. 124 Socialwall comments about the Socialwall experience ....................................... 125 Connecting the Research Findings ............................................................................... 126 Limitations ..................................................................................................................... 128 Limitations in design .............................................................................................. 128 Limitations in sampling ......................................................................................... 129 Limitations in procedures ...................................................................................... 130 Implications and Recommendations ............................................................................ 131 Implications and recommendations for practice ................................................. 132 Implications and recommendations for research ................................................ 133 Implications and recommendations for theory ................................................... 136 Conclusion ...................................................................................................................... 139 APPENDICES ................................................................................................................. 141 APPENDIX A: Call for Participation .................................................................. 142 APPENDIX B: Informed Consent ....................................................................... 143 ix APPENDIX C: Notice of Group Assignment - Control ...................................... 145 APPENDIX D: Notice of Group Assignment - Experimental ............................ 146 APPENDIX E: Pre -test ........................................................................................ 148 APPENDIX F: Pre -test Reminder - Control ....................................................... 153 APPENDIX G: Pre -test Reminder - Experimental ............................................. 154 APPENDIX H: Post -test Email ........................................................................... 155 APPENDIX I: Post -test ....................................................................................... 156 APPENDIX J: Post -test Reminder ...................................................................... 161 APPENDIX K: Interaction Prompts .................................................................... 162 APPENDIX L: Coding S chemas ......................................................................... 164 REFERENCES ................................................................................................................ 167 x LIST OF TABLES Table 1 Paired -samples t -tests comparing within -group pre - and post -test composite means ................................................................................................................................. 74 Table 2 Socialization construct group means for pre -test and post -test ........................... 76 Table 3 Themes emerging in Socialwall interactions ........................................................ 81 Table 4 Participant perceptions of the SocialwallÕs support for socialization ................. 89 Table 5 Themes emerging in post -test regarding Socialwall for socialization ................. 92 xi LIST OF FIGURES Figure 1 . Socialization via social media .......................................................................... 44 Figure 2 . Nonequivalent (pretest and posttest) control -group design .............................. 47 Figure 3. The Moodle Socialwall ..................................................................................... 55 1 CHAPTER I Introduction Adjunct faculty members are a growing resource in higher education. With financial models that more heavily rely on tuition -based funding in a time of nationally -decreasing enrollment (Perez -Pena, 2013), community colleges in particular are more often emp loying adjunct faculty members instead of adding to or filling full -time faculty appointments (Baldwin & Wawrzyinksi, 2011; Leslie & Gappa, 2002). The variable nature of adjunct faculty appointments allows community colleges to respond rapidly to fluctuat ing enrollment trends (Charlier & Williams, 2011). However, these appointments also bring instructional and organizational challenges such as concerns about instructional quality when adjunct faculty members replace full -time faculty (Baldwin & Wawrzyinsk i, 2011; Mueller, Mandernach, & Sanderson, 2013) and the stability of staffing (AFT, 2010). Organizational socialization, for this study, is defined as Ò...the extent to which newcomers (such as adjunct faculty) acquire the knowledge, skills, and function al understanding of their new jobs; make connection with others in the organization; and garner insight into the culture, processes, and people in their new organizationÓ (Bauer & Erdogan, 2014, p. 440). It is one factor that is strongly linked to employe e satisfaction, retention, and success (Bauer & Erdogan, 2014; Cooper -Thomas, Anderson, & Cash, 2011). Coupled with the adjunct faculty populationÕs vast range of experience, preparedness, schedules, and goals (AFT, 2010), there is a significant need for ongoing attention to organizational socialization methodologies that meet adjunct needs (Watts & Hammons, 2002). Previous research, for instance, found that the population of adjunct faculty central to this study experiences disconnectedness from 2 their co lleagues and from others within the institution (Wicks, 2013). How can we encourage adjuncts to participate in meaningful and structured socialization activities when they have such disparate needs, schedules, and levels of institutional connection? This research proposes social media as one potential means to address this question. Next, to help orient the reader, I present a definition of the key terms I will reference in this introductory chapter and throughout the dissertation. Definition of Key Ter minology Adjunct faculty. The terms part -time, adjunct, and contingent faculty members are often used interchangeably. These terms refer to faculty members that are non -tenure track, which will also serve as the definition of adjunct faculty members for t he purposes of this study. Non -tenure track faculty members have become the majority of faculty members in U.S. community colleges (Gappa, Austin, & Trice, 2007), and this transition to a non -tenure track appointment model is a major change in U.S. higher education. Institutions must develop models to support these part -time, adjunct faculty members. This support will then enable them to more effectively accomplish their own goals as educators and the goals of the institution in terms of employee and custo mer satisfaction, service quality, and staff longevity (Bauer & Erdogan, 2014; Cooper -Thomas et al., 2011). Social media. Kaplan and Haenlein (2010) describe social media as those highly -interactive , Internet -based applications that allow for user -generated content and exchange amongst groups and individuals, which will also serve as the definition of social media for this study. Examples of social media range from online social and professional net working platforms such as Facebook and LinkedIn, to blogging and 3 microblogging platforms such as Blogger and Twitter, to synchronous communication tools such as Google+ Hangouts and Skype, to social aggregation and sharing outlets such as Pinterest and You Tube. Common features of such platforms are the ability to post or share, comment or discuss, Òlike,Ó and ÒfollowÓ to build networks of those with similar interests and backgrounds. In this way, users are exposed to a constant stream of information, reso urces, Òexperts,Ó and dialogue related to their areas of interest along with a platform on which to share resources and perspectives for public feedback. Socialization. Socialization, often referred to more specifically in the literature as organization al socialization or onboarding , is one way of integrating part -time faculty members in a meaningful way into the life of the whole institution. As defined by Bauer and Erdogan (2014), organizational socialization is Ò...the extent to which newcomers acqui re the knowledge, skills, and functional understanding of their new jobs; make connections with others in the organization; and garner insight into the culture, processes, and people in their new organizationÓ (p. 440). The working definition of socializa tion for this study i s based on the work of Schrodt, Cawyer, and Sanders (2003), who identified three essential dimensions of socialization for new faculty in the communication discipline. These three dimensions of socialization are: 1) a sense of ownersh ip of the role, 2) receipt of adequate information to perform, and 3) a sense of connectedness to colleagues. Several examples of activities that may fit into this definition of socialization include orientations, mentoring, and ongoing opportunities for collaboration and professional development such as workshops, brown bag sessions, and so on. Most research on these often -simultaneous constructs has made use of survey instruments to gauge faculty member perception and satisfaction in these areas. 4 Resear ch Goals To explore the efficacy of social media as a low -cost solution for the onboarding, or socialization of adjunct faculty members, 77 adjunct faculty members at a small, Midwestern community college took part in an intervention study facilitated with a new social media format plug -in for the Moodle Learning Management System (LMS) called Socialwall. These adjuncts were randomly assigned to either an experimental or a control group and were asked to complete both a pre -test and post -test survey to hel p track changes in self -reported measures of socialization across time. Although the literature illustrates that social media has the potential to promote communication and collaboration; to support the formation of identity and knowledge development; to foster new forms of inquiry; and to impact students cognitively, socially, and emotionally (Pimmer, Linxen, & Grıhbiel, 2012; Ranieri, Manca, & Fini, 2012); the research on these tools in academic contexts has focused on their use as supplements to formal classroom learning experiences for students. The goals of this study were to add to the sparse literature in the areas of organizational socialization of adjunct faculty members and the use of social media to support ÔcommuterÕ audiences such as faculty a nd students at community colleges , and also to examine the potential of one specific tool to connect with, support, and develop an adjunct faculty community. Next in this introductory chapter, I present the problem my research seeks to address. I then present my purpose in conducting this research followed by my hypotheses and research questions and an overview of my methods. Then, I will discuss the rationale for, and significance of, the research followed by an exploration of my unique role in it and the assumptions that I bring into this research. My goal in this 5 introductory chapter is to give the reader an overview of this dissertation study before turnin g to an in -depth discussion of the relevant prior literature and theory in which my work is situated and my methodology, results, discussion, and conclusions in the chapters that follow. Problem Statement Hiring adjunct faculty to flexibly respond to enro llment trends is far more convenient and less costly than establishing more permanent, full -time faculty appointments. However, the adjunct faculty population has unique needs (Watts & Hammons, 2002). They are adult learners with varied degree attainment and they have a range of experience levels, often lacking any formal preparation for instructional success (AFT, 2010). They may also hold other commitments, dividing their time and attention. While some teach into retirement, others are career building , and others are simply freelancing or supplementing income. The literature suggests that adjunct faculty can be subject to burnout (Bates, 2012 ; Stout, 2013 ) and a sense of disconnection from their colleagues and their institution ( Jolley, Cross, & Bryan t, 2013; Meixner, Kruck, & Madden, 2010; Wicks, 2013). The literature also suggests that the teaching practices and performance of adjunct faculty can vary greatly from those of their full -time colleagues (Baldwin & Wawrzynski, 2011 ; Kezar, 2013 ), and tha t adjunct staffing may be correlated to less positive academic outcomes for students (Eagan & Jaeger, 2008; Jacoby, 2006; Umbach, 2007). Although some studies dispute the fact that using adjuncts negatively impacts student outcomes (Landrum, 2009; Ronco & Cahill, 2004; Thyer, Myers, & Nugent, 2011), the American Association of University Professors was concerned enough to author an official statement on how this hiring model can negatively impact 6 educational quality, collegiality, and academic freedom (AAU P, 2010). The part -time status of adjunct faculty has been characterized as Òone of the greatest challenges that community colleges faceÓ (Gonzales, 2009, p. 6). In the model of socialization framing this study, a broad range of concerns for adjunct staf fing (connection, instructional efficacy, a sense of support, etc.) may be best addressed by attending to three constructs that work in harmony to support a satisfying socialization experience: ownership , adequate information , and connectedness (Schrodt et al., 2003). Briefly, ownership refers to a sense of loyalty to the institution and feeling valued by it and the people in it. Having adequate information refers to receipt of information and resources sufficient to perform successfully in the role. Lastl y, connectedness refers to interactions and friendships with colleagues. Through literature, research, and practice, it seems that the socialization needs of adjunct faculty are not agreed upon, though they remain unmet by current, inconsistent or ineffective practices (Gillespie & Robertson, 2010; Langen, 2011; Meixner et al., 2010 ; Roueche, Roueche, & Milliron, 1996). The research on socialization in academic settings often looks at interventions such as the ment oring of junior faculty at four -year institutions (Alexander, 1992; Perna, Lerner, & Yura, 1995; Cawyer & Friedrich, 1998; Cawyer, Simonds, & Davis, 2010) or orientation initiatives (Nelson & Quick, 1991; Saks, 1994; Wanous, 1993). These are typically onl y marginally useful to adjunct faculty, especially those at community colleges, due to the very different appointment model in which they serve. Some institutions have initiated adjunct faculty coordinator positions or enhanced orientation and mentoring p rograms to socialize adjuncts and address the needs of this population, but solutions requiring staffing or program funds are particularly 7 difficult for small, enrollment -funded institutions such as community colleges to support. Very few institutions rep ort having formalized socialization efforts for adjunct faculty, and even fewer report being satisfied with the practices they employ for this aim (Roueche et al., 1996). Likewise, only around 63% of adjuncts receive performance evaluations from their adm inistration (Langen, 2011), despite adjunctsÕ interest in professional development to enhance their studentsÕ engagement and learning (Meixner et al., 2010). The need for low -cost, low -maintenance solutions that are flexible enough to address individual n eeds and omnipresent enough for those with a variety of other commitments is clear. Social media has the potential to meet this need. Research on social media has illustrated that it can support processes such as social learning (Robelia, Greenhow, & Burt on, 2011), identity development (Byrne, 2007; Davis, Deil -Amen, Rios -Aguilar, & Gonzalez Canche, 2012), professional information sharing (Gruzd, Staves, & Wilk, 2012; Rowlands, Nicholas, Russell, Canty, & Watkinson, 2011), the building of social capital (B urke, Marlow, & Lento, 2010; Ellison, Steinfield, & Lampe, 2007; Ellison, Wohn, & Greenhow, 2014), and an individualÕs receipt of emotional support (Birnie & Horvath, 2002; Gao, Luo, & Zhang, 2012) among other things. These processes are critical to owner ship, adequate information, and connectedness, the three constructs in the model introduced above for satisfying socialization experiences (Schrodt et al., 2003). This study aims to add to the sparse literature in the areas of organizational socialization of adjunct faculty members and the use of social media to support ÔcommuterÕ audiences such as faculty and students at community colleges. It also examines the potential of one specific tool to connect with, support, and develop an adjunct faculty commun ity. 8 Statement of Purpose Clearly, the hiring of adjuncts brings a variety of benefits in terms of institutional expenses, staffing flexibility, and employee experience, but it also brings many challenges in terms of employee and staffing commitment, ease of connection, and instructional quality. Organizational socialization is one factor shown to enhance the employee experience, support the success of employees in their roles, and promote lengthier retention of these employees at the institution. Further more, prior research has shown that adjunct faculty in a community college setting perceive several areas for potential enhancement of their socialization experience in terms of connectedness, including both opportunities to socialize with work colleagues and to formally interact with these colleagues during the work day (Wicks, 2013). In a literature review exploring the use of social media in educational contexts, Minocha (2009) asserted that social media could promote group interaction, allowing a low -barrier to participate in a community of inquiry, and facilitate feedback that increases participantsÕ motivation. I propose that the benefits of social media may then provide a viable solution to meeting many institutional and population needs for adjunct faculty members. In addition to exploring the efficacy of these tools for employee socialization, this study will provide one tool for administrators seeking to implement social media into their efforts in supporting the adjunct faculty member audience. Research Questions and Hypotheses In an attempt to explore the potential of social media (i.e., the Moodle Socialwall plug -in) for supporting adjunct faculty socialization, I posed the following research question along with several related hypotheses: 9 RQ1: Do adjunct faculty participating in the Moodle Socialwall intervention report higher perceptions of socialization than their peers? Related to this research question, I proposed three hypotheses: H1: Adjunct faculty participating in the Moodle Socialwall intervention will report higher average perceptions of socialization than their peers. H2: Adjunct faculty participating in the Moodle Socialwall intervention will report higher average perceptions of socialization following the intervention. H3: Adjunct faculty not participating in the Moodle Socialwall intervention will report consistent perceptions of socialization across time. Two additional research questions were posed to provide further context for the results: RQ2: What themes in dialogue and participation emerged during the experimental group intervention? RQ3: What benefits , if any, did the experimental group perceive that the intervention provided to support their socialization? In relation to these questions, I predict that emerging themes from intervention interactions will be related to the three constructs of socialization (ownership, adequate information, and connectedness) and that, in general, the experimental group will perceive that the intervention supported their socialization. Overview of methodology . The research methodology utilized in this study was a pre -test post -test control -group design. The 77 participants, all adjunct faculty members at a small, Midwestern community college, were randomly assigned to either a control gr oup or an experimental group. Both groups completed a pre -test survey and 10 then, after one month had passed, a post -test survey. The surveys were designed to capture their satisfaction with their socialization experience at the college as based on the thr ee construct model presented by Schrodt et al. (2003). As previously stated, the three constructs comprising a successful socialization experience in that model are ownership, adequate information, and connectedness. The experimental group was also asked to participate in an intervention utilizing the new Socialwall format plug -in for the Moodle LMS for the one -month duration of the study. The Socialwall format helps a Moodle course space function like a social networking site, with a ÒwallÓ or ÒfeedÓ of activity and the ability to ÒlikeÓ and ÒshareÓ with enrolled peers. The pre - and post -test surveys also gathered population demographics, asked of communication methods and patterns, explored social media use, and requested feedback from experimental gro up members on the intervention experience. Due to limitations in sample size, quantitative data were analyzed through descriptive statistics and a series of t -tests, while qualitative data were analyzed through content analysis. Rationale and significance . In a previous study of socialization specific to the adjunct faculty population described here (Wicks, 2013), adjuncts reported positive perceptions of socialization in terms of their sense of ownership and receipt of adequate information to perform in their roles, but reported mixed feelings on their connectedness to colleagues and others at the institution. Some participants also perceived that they did not have sufficient opportunities to formally interact with colleagues during work nor did they hav e sufficient opportunities to socialize with colleagues when not at work. Another intriguing trend expressed in this survey was that long -serving adjunct -participants reported perceptions that they were less adequately informed for their role and recently 11 hired adjunct -participants reported perceptions that they were more adequately informed for the role. One potential explanation for this surprising finding was that newly formalized hiring and mentoring practices were put in place within the past 5 years in the community college, along with two attempts at a (now defunct) adjunct faculty coordinator position. These recent initiatives could have impacted newer hiresÕ more positive perceptions of being adequately informed for their roles compared to their veteran peers. Moreover, qualitative data from the survey suggested that members of this adjunct faculty population felt undervalued and unheard and that some of these adjuncts had perceptions that college administrators often asked for feedback but did no t apply it in practice. This study was designed to provide a solution to address some of the needs expressed by this population. In addition to informing adjunct faculty onboarding initiatives at this community college site, the research described here will contribute to the literature related to the use of social media in education, broadening its scope from a primary focus on student learning to focus on its potential for the professional dev elopment of educators. The intervention designed herein could also provide one tool for administrators seeking to consider low -cost, low -maintenance solutions for adjunct faculty onboarding Role of the researcher . I have served as the primary researcher in terms of planning and conducting the study. However, I have received input through my dissertation committee and the Center for Statistical Training and Consulting (CSTAT) at Michigan State University. I also served as a facilitator for the intervent ion, posting 1 -2 conversation starters each week during the one month of its duration. Though this role and these interactions may have some impact on participant perceptions of the 12 experience, it is clear that some level of interaction with others is nec essary in the framework for socialization presented by Schrodt et al. (2003). For these reasons, it is necessary to provide a baseline level of interaction in the Socialwall space. Though many adjuncts were aware of my role as an administrator at the ins titution, I do not serve in a supervisory capacity to them, and they are also aware of my dual role as an adjunct faculty member at the institution. These facts, I believe, assisted their perceptions of the authenticity of our interactions. A more formali zed approach, attempting to meet more formal learning objectives or agendas through the intervention, would stray more toward formal intervention design rather than toward the focus of this study, which was the more informal potential of the Socialwall too l to organically impact socialization. Researcher assumptions . Inherent in this research is the assumption that increased connectedness to colleagues and institutionally initiated workplace socialization tactics are ideals worthy of pursuing as an insti tution. Though the organizational socialization literature and research illustrates that institutionalized socialization efforts increase retention and success in the role (Ashforth, Sluss, & Saks, 2007; Bauer, Bodner, Erdogan, Truxillo, & Tucker, 2011), it should be noted that several members of this adjunct population indicated in prior research that they do not view connecting with their colleagues as a valuable activity that will assist their performance and also that they do not wish to pursue related interactions (Wicks, 2013). Additionally, due to the enormous number of global social media users, this research assumes that engagement with others through social media is desired by many and somewhat inevitable to our professional futures as educators, and also that social media is a suitable intervention to meet the socialization needs of most adjunct faculty. However, it should 13 also be noted that several members of this adjunct population were also very direct about the fact that they view technology -facilitated communications , such as those occurring on social media, as impersonal, inferior alternatives to face -to-face communications , and they do not wish to engage with others through these tools (Wicks, 2013). Specific attitudes and experiences I have had that could impact the validity and reliability of this research design include my work as an educational technologist, instructional designer, and a director of distance learning. In this role, I have been ask ed to assist in forging innovative, effective solutions for adjunct faculty orientation and development at the small, Midwestern community college where this research was conducted. The instructional staff of this community college is largely comprised of adjunct faculty members with varied experience levels and schedules, and the institution itself is mainly tuition -funded, seeing a decrease in overall enrollment in the last several years. However, enrollment in distance learning opportunities holds stea dy during this time and adjunct faculty participation and interest in the institution's distance -deployed Teaching Online Core Competency Credential and other online professional development opportunities remains strong. As an individual geared toward ind ependent learning experiences as facilitated by technology, I find immeasurable value in use of Rich Site Summary (RSS) aggregators and social media outlets. These serve as a means to enhance my professional development and sense of connection to the comm unity through resource sharing, interactions with others, and exposure to new approaches and perspectives. Therefore, I am compelled to share my passion for this style of connection and self -directed learning with others. I believe that if I can illustra te the ease and benefits of participation in these environments, even simply as consumers of information 14 shared by others, most individuals would have a similar galvanizing experience upon participation. Without a doubt, this impacts all aspects of this r esearch, from the selection of the focal topic, to the phrasing of the research questions, to the methods and the manner in which results are reported. For example, I may be likely to overemphasize results that provide evidence of support for socialization in data analysis and underemphasize those that do not. I will safeguard for these impacts by exploring multiple data sources and seeking out discrepant data, among other tactics. Organization of the Dissertation In summary, the socialization needs of adjuncts are unmet, and this may have notable implications for adjunctsÕ practices and retention. Social media may provide a low -cost tool to enhance socialization for this audience. In Chapter II, I will provide a focused review of the literature relate d to these topics, including research on the organizational socialization of faculty and social media use in higher education. I will also address the limitations of work in those arenas, illustrating how the existing research and its gaps provide a conce ptual frame for this study. Then, in Chapter III, I will outline my methodology. This will include details on the setting and sampling, the instrumentation, procedures, and analysis. Along with this, I provide a rationale for these choices in context wi th the hypotheses and research questions. In Chapter IV, I will move on to research findings, discussing the analysis and synthesis in detail. That discussion will provide depth in interpretation, issues of validity and reliability, and limitations. Fina lly, in Chapter V, I provide conclusions and recommendations derived from the data, posing questions for future research. 15 CHAPTER II Literature Review One of the goals of this study was to contribute to the literature related to the use of social media f or faculty development. Specifically, I sought to explore the efficacy of one social media tool, the new Moodle LMS Socialwall format plug -in, for the use of adjunct faculty socialization (i.e., ownership, information sharing, connection). In this study, each of the three constructs in this model of socialization (described in more detail in the next section) was evaluated through participant self -report on Likert -type survey questions. Ownership , the first construct of socialization in this model, is re presented through questions regarding oneÕs likelihood to apply to their department if engaged in a job search, a sense of being valued in the work environment, a feeling of ownership of the department, and a sense of loyalty to the department. Adequate i nformation , the second construct of socialization, is represented through questions regarding the receipt of adequate information to serve in oneÕs job role , the access to adequate resources to serve, clear explanations of departmental service expectations, and the participantÕs understanding of expectations for continued employment at the institution. Connectedness , the third and final construct for sociali zation in this model, is represented through questions regarding a feeling of connection to others in the work environment, opportunities for socialization with colleagues, opportunities for formal interactions with colleagues during work, and the sense th at oneÕs colleagues are also considered to be friends. To situate this study in the context of the literature related to organizational socialization for faculty as well as the literature related to social media in higher 16 education, I next provide a n overv iew of existing work in these two areas. This review will illustrate both the need for focused attention on the design of flexible socialization experiences that better meet the diverse needs and schedules of adjunct faculty and also the need for further research on social media as a tool for educatorsÕ professional development. Themes, methods, limitations, and conclusions from reviewed works are discussed. These works were pulled from a variety of databases and peer -reviewed journals using search terms such as: organizational socialization, socialization, onboarding, faculty, adjunct faculty, social media, social networking, and Facebook. Because of the emphasis on current developments in adjunct hiring and support and the rapid developments in social media tools between 2006 and 2014, my search of various educational databases and journal contents was largely limited to these past 8 years. Additionally, the sources referenced here are primarily those related to higher education and faculty. Though I recognize that socialization and social media research may extend into many disciplines, it is most fitting to explore the pieces more directly related to this population and this phenomenon. I close this chapter by presenting the conceptual framework for this study. Organizational Socialization Organizational socialization has been explored in the literature for at least 40 years as a critical component of employee performance, satisfaction, and retention along with overall organizational effectiveness (B auer & Erdogan, 2014; Cooper -Thomas et al., 2011; Saks, Uggerslev, & Fassina, 2007). Effective approaches can lead to rapid learning, task mastery, social integration, job satisfaction, organizational commitment, retention, and a variety of other factors (Cooper -Thomas et al., 2011; Kammeyer - 17 Mueller, Livingston, & Liao, 2005; Kramer, Callister, & Turban, 1995; Ostroff & Kozlowski, 1992). Fang, Duffy, & Shaw (2011) outlined three tracks of existing research in the field of organizational socialization: (1) organizational methods and tactics for newcomers, (2) information acquisition interactions between the individual and the environment, and (3) the experience and characteristics unique to individual newcomers. As the goal of this study is to explore a to ol that may support organizational socialization tactics, my discussion of the literature in this area will focus on the organizational track. Faculty socialization tactics. According to Fang et al. (2011), ÒAn organizational approach involves examining m ethods and processes (e.g., organizational socialization tactics) that organizations use to structure newcomersÕ socialization experiencesÓ (p. 128). There has been a great deal of focus on orientations (Nelson & Quick, 1991; Saks, 1994; Wanous, 1993) and mentoring (Alexander, 1992; Perna et al., 1995; Cawyer et al., 2010) as tactics in the higher education and faculty socialization literature, though the specific role of these components in the socialization process is not yet well understood. Mentoring has been linked to the formation of interpersonal bonds and the receipt of advice and support from experienced colleagues (Cawyer et al., 2010). It has also been linked to increased newcomer knowledge about the organization and the newcomerÕs sense of com fort with the environment (Ostroff & Kozlowski, 1993). Adjunct certification programs (Burnstad, 2002; Scott, 2010; Zimmerman & Struthers, 2011) have also gained traction as of late. Though promising, these approaches can require a high degree of staff t ime and energy in terms of coordination. In the paragraphs below I describe each of these three tactics for organizational socialization (orientations, mentoring, certification programs), including evidence -based exemplars before turning to 18 their limitatio ns for the socialization of adjunct faculty members at community colleges and my resulting conceptual framework. Orientations. It has been reported that the quantity of time spent in orientation activities is likely one of the best predictor s of satisfaction upon arrival in the work force (Cawyer & Friedrich, 1998). To assist in describing what orientation might entail, I outline several instances of this tactic in practice as described in the research literature. First, Welch (2002) describe d a complex , year long, three -part orientation model employed at St. Louis Community College with new faculty members. It consists of an intensive week of orientation at the start of an academic year, followed by weekly, face -to-face sessions during the f all, and an instructional skills workshop during the spring/summer. The four proposed outcomes for the program are: modeling a learning -centered environment, enabling knowledge about the campus and its programs and services, encouraging collaboration and interaction among faculty and staff, and establishing an expectation for ongoing professional development. Though this model has proven very successful through extensive evaluations and participation since 1999, it is geared at full -time faculty members, p roviding compensation and release time for participation that may not be feasible with an adjunct faculty audience. Additionally, it requires continued, devoted, face -to-face time from faculty facilitators that are also provided release time from teaching to coordinate the experience. This may also pose an issue for institutions that already have strict staffing constraints due to enrollment and budgetary issues. Second, West, Borden, Bermudez, Hanson -Zalot, Amorim, & Marmion (2009) recommended a multi -faceted approach to orientation including a combination of a 19 workshop, collaboration with relevant outside agencies, mentoring from full -time faculty, and incorporation of technology for ready access to information to bolster the confidence and development o f adjunct faculty members. Their one -day workshop prior to the start of the academic year also allowed participants to receive continuing education credit, motivating greater participation. However, they found that adjunct faculty participants requested weekend and evening opportunities to address otherwise constrained schedules. At Johnson County Community College (JCCC) in Kansas, orientation emphasizes teaching and learning and is followed by four to five days of in -service activities at the start of the semester (Burnstad, 2002). Faculty members participate in all-staff meetings, meetings about special initiatives such as the Honors Program, select from concurrent sessions on teaching and learning topics, and can enroll in two - or three -day workshop s on topics such as the learning management system, Web technologies, or a master teacherÕs workshop. However, this again describes a model that requires a great deal of time, staffing, financial, and other resources. As these studies indicate, despite in terest from administration in dynamic orientation programs, they can be complex to administer, requiring a high -level of involvement from staff. Additionally, research and employee feedback related to orientation programs and events have reported mixed re sults in terms of the perceived utility and impact of such orientations on practice (Nelson & Quick, 1991; Saks, 1994; Wanous, 1993). Mentoring. A second organizational socialization tactic is mentoring. Schrodt et al. (2003) found in a national survey on socialization and mentoring among communication faculty that mentored faculty members feel more connected to their 20 institutions, have a greater sense of ownership over their roles, and believe they receive more information about organizational expectation s, opportunities, and practices than their non -mentored peers. Hessler & Ritchie (2006) indicated that they did not see formal evidence -based support in the literature for mentoring, yet they relied on personal experience and anecdotal evidence from a var iety of novice faculty members to deem it a priority in faculty retention. To further explain the concept of mentoring, I describe one exemplary mentoring program below. Lesley University in Cambridge, Massachusetts, has outlined an evidence -based, multi -stage Adjunct Mentoring Program that assists university administrators in supporting the growing number of adjunct faculty members they employ (Ziegler & Reiff, 2006). The process begins with the selection of faculty members that have both the skill and the interest required to serve as a mentor. Then, these mentor -prospects attend a day -long, interactive session with the program director to inform them of the program goals: (1) communicating essential course content, (2) modeling effective teaching, and (3) facilitating professional collaboration. Following these sessions, new adjunct faculty members are invited to take part in an intensive, three -day weekend orientation meeting with various stakeholders and the mentors. Following the orientation, adju ncts are assigned either to instruct a course, working closely with their mentor, or to serve as a teacherÕs assistant for a course under their mentorÕs guidance, if they have little prior experience with the university. Finally, as adjuncts become more comfortable with their teaching roles, the adjunct faculty member and the mentor keep up regular correspondence via email and/or a mentoring website about the evolution of the course, which assists both parties. 21 Though impressive in scope and based on years of fine -tuning related to em erging empirical literature (Zie gler & Reiff, 2004; Ziegler & Reiff, 2005; Ziegler & Reiff, 2006), a formalized mentoring program like that at Lesley may not be feasible for a community college with more economic and staffing c onstraints. Though the Lesley program is unique to adjunct faculty, mentoring research in academic settings often looks at less concrete interventions that are aimed toward junior faculty at four -year institutions (Alexander, 1992; Perna et al., 1995; Caw yer & Friedrich, 1998; Cawyer et al., 2010). Such research may or may not be applicable to community college adjunct faculty, who serve in a drastically different appointment model. Scott (2010) reported that only about half of community colleges have an y sort of mentoring effort in place. Adjunct certification programs. A third approach to organizational socialization is the adjunct certification program. Larcara (2010), for instance, discussed the need for modular training opportunities that can be s elf-paced and flexible to meet individual adjunct faculty member needs. She notes that though administrators have a preference for face -to-face approaches in practice, digital approaches might save institutions both time and money, as they could be self -paced or deployed by a single facilitator to meet demands. Additionally, the training modules can be altered and developed to meet fluctuating needs. To help portray the range of approaches for adjunct certification tactics, I here describe two more stand ard, documented models. Baron -Nixon (2007) outlined a required teaching certification for adjunct faculty members at Florida International University that spans the course of two semesters. Adjuncts must attend a series of face -to-face workshops related t o teaching, maintain a journal of their teaching experiences, write reviews of scholarly articles in their 22 discipline, and author a personal philosophy of education. The activities for ongoing professional development, personal reflection, and building a l earning community of oneÕs peers in Baron -NixonÕs (2007) book are informed by a great deal of empirical research. However, they also seem a lot to ask from part -time employees with complex life demands who are compensated at just a fraction of the amount that full -time faculty members receive for similar instructional activities. Another adjunct certification program is that of Johnson County Community College (Burnstad, 2002). Applicants must get support from their assistant deans and author statements o f intended learning outcomes. Participants must complete seven required modules and submit a reflective journal to receive a certificate of completion, an additional step on the salary schedule, and a book about teaching and learning. The seven modules i nclude: (1) orientation and narrative reflections, (2) employment policies and procedures, (3) technology, (4) designing effective instruction, (5) challenges of students, (6) legal issues and diversity, and (7) microteaching and videotaping of a classroom activity. Participants can also opt into one of a list of electives including options like communication, test construction, portfolio preparation, and so on. Burnstad (2002) found that in this model, the salary step increase motivates participation. However, financial incentives can be problematic for institutions seeking low -cost solutions. In closing, adjunct certification programs exhibit some of the same weaknesses as orientation and mentoring do as approaches to socializing adjunct faculty members. Depending on how these programs are executed, they may require a significant amount of coo rdinating, staffing resources, or financial motivators from already financially -strapped 23 institutions. After all, a major reason that community colleges leverage adjunct faculty for instruction is to offset the costs of operation in a time of uncertain tu ition and state revenues. Limitations in socialization research. The literature on organizational socialization in general is still a bit disconnected (Bauer et al., 2011). For instance, it was not until the late -1980s to the mid -1990s that researchers attempted to integrate the research on training into that on socialization (Saks & Ashforth, 1997). Also, weak sampling procedures are of concern, as trends with new employees making school -to-work transitions may not match the trends of new employees making job -to-job transitions, and convenience samples often cannot be generalized to broader populations (Bauer et al., 2011). Additionally, much research on socialization gathers data throug h self-reports with survey and interview protocols that rely on accurate participant recall and report (Saks et al., 2007). However, it is apparent that organizational socialization is critical to organizational effectiveness (Louis, 1980), predicting lon g-term success of new employees (Ashforth et al., 2007), along with promoting their retention and increasing the likelihood that they will reach first -year performance goals (Bauer et al., 2011). Yet, studies suggest that very few college administrators a re devoting the time and attention needed to implement systematic socialization procedures for part -time faculty (Roueche et al., 1996; Shannon, 2007). In terms of socialization efforts specific to higher education, the research usually centers on full -time faculty newcomers, typically entering or seeking tenured positions, often in the communications discipline (Cawyer & Friedrich, 1998; Cawyer et al., 2010; McCoy, 2006; Schrodt et al., 2003; Tierney & Rhoads, 1994; Trowler & Knight, 1999). 24 However, with increasing reliance on adjunct faculty in times of fluctuating enrollment in higher education, some articles have focused on the diverse needs of the adjunct audience and current approaches to meeting them (Finucane & Algren, 1997; Roueche et al., 1996; Shannon, 2007). As described above formalized socialization tactics tend to offer adjuncts traditional, face -to-face options like orientations or mentoring, which can require a great deal of college resources and which may conflict with adjunct schedules or other obligations (Shannon, 2007). As a result, adjuncts report a sense of isolation, along with an unmet desire to connect with peers, and concerns that the lack of professional development opportunities available will negatively impact their instruction al quality and opportunities for advancement (Finucane & Algren, 1997). Though, it should also be noted that adjuncts with more extensive classroom experience are generally more satisfied with the socialization options available to them (Finucane & Algren , 1997). Conclusions. In summary, institutionalized tactics for formalized organizational socialization experiences tend to promote higher rates of employee success than do approaches that rely on individuals to socialize independently and informally to t heir new roles (Bauer et al., 2011; Cooper -Thomas et al., 2011; Saks et al., 2007). Though adjuncts crave socialization opportunities, there is a lack of formalized onboarding initiatives for this audience that meet their schedule and socialization needs ( Roueche et al., 1996; Shannon, 2007). Tactics tend to include orientations or mentoring, with adjunct certification programs also becoming a popular option. However, these approaches are often involved and costly in terms of institutional time, staffing, and financial resources. In addition, the organizational socialization research is limited in terms of sampling (Bauer et al., 2011), as much of the work in this area focuses on new tenure - 25 track faculty in communications -related fields (Cawyer & Friedri ch, 1998; Cawyer et al., 2010; McCoy, 2006; Schrodt et al., 2003; Tierney & Rhoads, 1994; Trowler & Knight, 1999). Adjunct faculty, when surveyed, report a sense of isolation (Finucane & Algren, 2007; Shannon, 2007; Wicks, 2013) and a desire for relevant professional development opportunities (Finucane & Algren, 2007; Wicks, 2013). Specific to the faculty development that occurs at community colleges, Murray (2001) suggested that it lacks goals, lacks evaluation, and fails to reach those with greatest nee d for development. In looking at the constructs of socialization for this study: ownership , adequate information , and connectedness , formalized organizational socialization efforts have the potential to support these needs by providing opportunities for re levant information sharing, role clarification and validation, along with connection with others at the institution. Though popular approaches sometimes fail to meet these needs and require much in terms of coordination and other resources, social media o ffers unique affordances that may readily facilitate flexible connections and development at a low cost. Social Media Now, I turn to the literature related to social media use in higher education. As mentioned in chapter I, social media is defined as hig hly -interactive, Internet -based applications that allow for user -generated content and exchange amongst groups and individuals (Kaplan & Haenlein, 2010). Facebook is the most widely investigated tool due to its current cross -cultural and cross -demographic pervasiveness (Tess, 2013). Though it has been claimed that social media may enhance teaching and learning (Minocha, 2009), much of the research in this area has mixed results, likely because it is 26 based on self -reports and individual perceptions as oppo sed to formal observations of teaching and learning and student outcome data (Davis et al., 2012; Tess, 2013). The research that does exist within academic contexts suggests that the integration of social media into teaching and learning can promote commu nication and collaboration, identity development, knowledge development, and that it can also affect students cognitively, socially, and emotionally (Pimmer et al., 2012; Ranieri et al., 2012), though most of these studies have been executed at 4 -year, res idential institutions (Davis et al., 2012). There is a particular scarcity of social media research in the community college context, where it might provide essential access to services and to peers for commuter students and commuting adjunct faculty. In a recent survey, only 13% of community colleges reported any substantial, institutionally -driven use of social media (Davis et al., 2012). Here, I outline some of the themes in the educational research literature related to the use of social media in an academic context. Social media in teaching and learning . Many researchers have attempted to study social media use in relation to academic outcomes and engagement in higher education. Although some research has illustrated significantly lower grade point averages for college students who are Facebook users than for non -users (Kirschner & Karpinski, 2010), other research has shown increased grade outcomes (Junco, Heibergert, & Loken, 2011; Paul, Baker, & Cochran, 2012). This may be due to the fact that so cial media can enhance student -faculty connection, facilitating information sharing including clarified expectations and prompt feedback, while also providing opportunities for connection through collaborative learning and ownership of the content and educ ational experience through active learning (Davis et al., 2012). Several studies have found that 27 the addition of social media to a course experience, or the use of a social media channel (e.g., Facebook or Twitter) in place of a traditional learning manag ement system in the course, consistently increased the volume of complex communications between students (Brady , Holcomb, & Smith , 2010; Schroeder & Greenbowe, 2009). This may be because the traditional class forums compared in these studies yielded few responses, while social media yielded more responses more quickly, which prompts increased frequency and variety of conversations related to course topics. These again illustrate the support of social media for information sharing and connection and hint at ownership, as those receiving many responses to inquiries may feel more valued. Additionally, the integration of social media into course experiences has been linked to deeper interpersonal connections and positive psychosocial development for students, some of the reasons being that social media provides both an easy access point for connectedness through peer interactions and connectedness because it provides an opportunity for participants to receive social support from, and coordinate additional inte ractions with, their peers (events, text messaging, etc.) that transcend the online space (Heiberger & Harper, 2008; HERI, 2007; Junco et al., 2011). Though some research suggests that social networking may increase anxiety in those with low social compete nce (Rorhle & Sommer, 1994), others have focused on its links to increased social support and stress reduction (Birnie & Horvath, 2002; Valkenburg, Peter, & Schouten, 2006; Wright, 1999). For example, in a study of Dutch adolescents using a regional socia l networking site, the frequency of use had indirect effects on adolescentsÕ self -esteem and well-being, and positive feedback from peers was linked to increased self -esteem and 28 well-being (Valkenburg et al., 2006). Though it should be noted that, in this study, negative feedback from peers had the opposite effect. Social media in scholarship and identity development. With social media outlets like Facebook currently at 1.31 billion monthly active users (Statistic Brain, 2014a) and Twitter at almost 646 m illion active users (Statistic Brain, 2014b), the statistics demonstrate that people are naturally opting in for a variety of reasons. Additionally, the literature, although scant, suggests that academics are following the same trend, and they are capital izing on the benefits of social media not only for personal connections but also for professional development and professional connections (Gruzd et al., 2012; Moran, Seaman, & Tinti -Kane, 2011; Rowlands et al., 2011). Gruzd et al. (2012), for instance, i n studying faculty acceptance and use of technology, found that the following trends promote faculty use of social media: (1) assumptions that it will enhance their performance, (2) impressions that it is easy to use, (3) feelings that the organization and technical structure will support them in their use, and (4) assumptions that others think they should make use of the tools. Scholars are turning to social media tools professionally because they are perceived as more convenient for making new connection s with peers and for collaboration and research dissemination (Gruzd et al., 2012). Whether used professionally or personally, social media can be beneficial for the development and expression of identity (Grasmuch, Martin, & Zhao, 2009; Greenhow & Robeli a, 2009; Pempek, Yermolayeva, & Calvert, 2009; Subrahmanyan, Reich, Waechter, & Espinoza, 2008; West, Lewis, & Currie, 2009). This is because it provides a semi -public forum for sharing preferences, characteristics, interests, and affiliations and 29 some re search suggests that this sharing may help enrich relationships in ways that support identity development (Davis et al., 2012). Social media in social and civic endeavors. Researchers have also studied the impact of social media on social capital and civi c engagement. While some research suggests that increased time online with social media can inhibit offline engagement, this research also indicates that how the time is spent online is linked to offline outcomes (Junco, 2012). For instance, those engage d in direct online communication activities with others, such as messages back and forth with contacts, were more likely to be engaged with others offline, while those involved in passive communications online, such as viewing and ÒlikeÓ -ing othersÕ posts, were less likely to be engaged with others offline (Junco, 2012). Also, several studies have linked social media to civic engagement, illustrating that social media provides an awareness of diverse perspectives, opportunities for peer modeling, and can p rovide space for a call to action that can motivate behavior offline (Beach & Doerr -Stevens, 2011; Greenhow, 2010; Robelia et al., 2011). In terms of social capital, or the information or social supports available to people via their social networks, socia l media has been shown to support both bonding and bridging social capital. Bonding social capital refers to resources or supports found within peopleÕs existing tight -knit groups (close ties) and bridging social capital refers to resources or supports wit hin groups of diverse others (loose ties) (Burke et al., 2010; Ellison et al., 2007; Ellison et al., 2014; Greenhow & Burton, 2011; Valenzuela, Park, & Kee, 2009). In a study of 286 undergraduate students, more intense Facebook use was linked to higher le vels of self -esteem and life satisfaction and higher bridging social capital (Ellison et al., 2007). Additionally, there was an association between lower rates 30 of social capital and lower scores of well -being for these students (Ellison et al., 2007). Also, in a study of 43 first -generation college students in the Midwest, some online interactions on Facebook were linked to increased incidence of bridging social capital, which exposed them to new possible careers, cultures, and life paths (Ellison et al., 2014). In another study, first -generation students who were connected on Facebook to someone that had enrolled in or graduated from college were more confident that they could attend and graduate from college (Wright, 1999). Conclusions. Facebook is t he most frequently studied social media (Tess, 201 3) and, in the realm of education, this research if often focused on the self -reports of 4 -year university students (Davis et al., 2012). However, in looking at the constructs of socialization for this stu dy (i.e., ownership, adequate information, and connectedness) social media has a noted potential to support these needs in some audiences. For instance, social media has the capacity to meet information needs by facilitating information and resource shari ng, along with providing spaces and tools for peer collaboration (Brady et al., 2010; Gruzd et al., 2012; Schroeder & Greenbowe, 2009). Additionally, social media can scaffold identity development by allowing for professional development and connections a long with providing a forum for expressing oneÕs preferences, interests, and affiliations (Grasmuch et al., 2009; Greenhow & Robelia, 2009; Moran et al., 2011; Pempek et al., 2009; Rowlands et al., 2011). Likewise, social media may promote connectedness t hrough its ability to help users form both bonding and bridging social capital and the ability to motivate civic behaviors (Beach & Doerr -Stevens, 2011; Ellison et al., 2014; Greenhow & Burton, 2011; Robelia et al., 2011). 31 Though some have posited that onl ine communication networks are more often used by lonely and socially -anxious individuals to establish relationships (McKenna & Bargh, 2000), or that social media can jeopardize traditional forms of social engagement (Kuh, 2009), research shows that online communication can be incredibly social, intimate, and friendly (Walther & Parks, 2002). Many people do use social networks to maintain relationships with individuals that are already part of their offline social networks (Kay, 2007), but these technologi es also support increased bridging social capital, or diverse networks of loose ties, which may expose people to diverse perspectives and new social connections (Ellison et al., 2007; Greenhow & Burton, 2011; Valenzuela et al., 2009). Social networks may h elp users develop and express identities or learn new skills and behavior (Gao et al., 2012; Greenhow & Robelia, 2009; Ito et al., 2008). But, there is a need for more data outside of self -report in the research. There is also a particular need for resea rch on the use of social media with the community college audience and for the benefits of adjunct faculty, a commuter population that could directly benefit from these affordances. Much of the literature reviewed focuses on student use of social media, which is a very different context from institutionally -sponsored use of social media with adjunct faculty. However, this research is meritorious as a means to explore if the benefits of social media evidenced in other contexts have the potential to manifest in this context as well. Justification for This Research As evidenced in the proceeding paragraphs, in terms of organizational socialization, there is a scarcity of empirical study related to the onboarding of adjunct faculty, particularly those at community colleges. Most of that work is focused on the 32 socialization of new faculty in tenure -track positions at four -year institutions. Though there is growing attention in higher education to strategies for developing the adjunct faculty body, it i s not often situated in the context of the community college. Likewise, in terms of social media, there is a shortage of empirical literature related both to social media use for educatorsÕ professional development and to the utility of social media in community college environments. The related bodies of literature are mostly focused on social media as a supplement to student learning in four -year institutions, though there is emerging literature addressing the social media usage habits of full -time facu lty at four -year institutions. Thus, more evidence -based socialization solutions are needed for adjunct faculty that are also flexible in terms of their content and delivery to accommodate adjunct faculty membersÕ varying schedules and needs. The studies described above also suggest how social media could help meet this need, with its affordances for facilitating and enhancing interpersonal connections and collaborations both inside and outside of the social media space, providing a forum for social suppo rt, aiding identity development, providing exposure to diverse perspectives and information, and promoting peer modeling that can prompt civic engagement and offline action. Thus, this study expands both the literature related to the socialization of adjun ct faculty and the use of social media for educator professional development. It presents one institutionalized socialization tactic that seems absent from the literature and it taps into the unique needs of community colleges in supporting their broad ad junct faculty audiences. 33 Conceptual Framework To re -state the issue at hand, adjunct faculty have a clear need for socialization that is often unmet. Additionally, the institutions serving adjuncts, particularly community colleges, struggle to find the financial and staffing resources to sustain socialization efforts for this audience. So, what was ascertained from the review of literature? Administrators should pursue those formal, institutionalized socialization efforts to promote effective practice s, successful integration, and retention of our adjunct faculty. Administrators should recognize the social capital embedded in adjunct faculty networks, pursuing socialization opportunities that are flexible in terms of scheduling and in content to bette r meet adjunct needs, and allowing them the opportunity to be contributors to the college community. Administrators could capitalize on the vast affordances of social media to meet these needs in terms of its flexibility, pervasiveness, and cost benefits coupled with its abilities to promote peer connection and collaboration, identity development, access to and sharing of information, and so on. It seems an ideal tool both for the needs of this audience and for the needs of the institutions serving this audience. In fact, there is a gap in the literature related to exploring socialization efforts and the potential of social media in relation to community colleges. This adjunct population, in particular, has a need for enhanced opportunities for connected ness (Wicks, 2013), which is one of the constructs of socialization in the model utilized for this study. In the following paragraphs I describe my conceptual framework for this study, specifically outlining how and why I have connected adjunct organizati onal socialization needs with social media affordances. 34 Thinking about organizational socialization. As discussed, organizational socialization has been studied through many lenses. Work in this area often references either descriptive models for sociali zation efforts, such as that by Van Maanen and Schein (1979) or large -scale theoretical models such as BanduraÕs (1986) conceptualization of social cognitive theory. Instead of exploring models that exist outside of higher education and how I might employ them within that context, my literature review focused on exploring frames that had illustrated utility within the unique context of faculty development in higher education. In their efforts to fill the gap on how socialization occurs in higher education, Cawyer & Friedrich (1998) engaged in a survey study of new faculty in the communications discipline. They found, as mentioned earlier, that orientation was key to communication faculty membersÕ satisfaction. They also identified differences between facu lty membersÕ perceptions of socialization during the job interview and when their work actually began. Their recommendations to administrators for enhancing faculty membersÕ socialization included clear definitions of daily work responsibilities, opportun ities for interpersonal work interactions, and exploration of mentoring initiatives. However, Cawyer & Friedrich (1998) provided no concrete theoretical model for socialization. Building on their work, Schrodt et al. (2003) explored the efficacy of mento ring activities on socialization in terms of benefits for newcomers and the institution. Through a principal components factor analysis, they found a three -factor model that accounted for over 65% of the variance for socialization. The three factors were : ownership, adequate information, and connectedness, which were then summed to represent satisfaction with the academic socialization process for participants. 35 Recall that these three constructs of socialization are loosely defined through related questio ns on a Likert -type scale. Ownership is represented by the likelihood of applying to the department if engaged in a job search, a sense of being valued in the work environment, a feeling of ownership of the department, and a sense of loyalty to the depart ment. Adequate information is represented by the receipt of adequate information to serve, access to adequate resources to serve, clear explanations of departmental service expectations, and an understanding of expectations for continued employment at the institution. Lastly, connectedness is represented by a feeling of connection to others in the work environment, opportunities for socialization with colleagues, opportunities for formal interactions with colleagues during work, and a sense that oneÕs coll eagues are also considered to be friends. In looking at three popular tactics for the socialization of new adjunct faculty in higher education (orientation, mentoring, and adjunct certification programs), it is evident that these needs are unmet by curren t approaches. For instance, if ownership is partially represented by a sense of being value in the workplace and adjuncts report feeling undervalued and unheard (Wicks, 2013), there is an unmet need. Additionally, if adequate information is partially rep resented by access to adequate information to serve and adjuncts report a need for further professional development opportunities (Finucane & Algren, 1997) or report that opportunities do not align with schedule availability (Roueche et al., 1996; Shannon, 2007; West et al., 2009), there is an unmet need. Likewise, if connectedness is partially represented by a feeling of connection to others and adjuncts report feeling isolated in their work (Finucane & Algren, 1997; Shannon, 2007; Wicks, 2013), there is an unmet need. 36 These three constructs identified in the Schrodt et al. (2003) model of socialization aligned closely with challenges that my institution and other community colleges have observed with adjunct staffing. These challenges include: reports of adjuncts feeling isolated, a lack of adjunct awareness of college policies and services, adjunctsÕ and administratorsÕ concerns with adjunct instructional effectiveness, adjunct disengagement from the courses, programs, or institution, and so on. To expl ore the efficacy of the Schrodt et al. (2003) model with this population, in my prior research, I slightly adapted their scale, identifying 12 questions with four geared to measure each of the three constructs of socialization. I then deployed that survey with the population (Wicks, 2013). The ChronbachÕs Alpha score for this scale from that research calculated at .903, suggesting a high level of reliability for the scale items in this context, which also aligned with participantsÕ comments. For these re asons, I find it a useful model in moving forward with research to aid this audience. Connecting organizational socialization to social media. The unmet socialization needs associated with current models of socialization for adjunct faculty include items such as opportunities to connect with administrators and peers (or connectedness ), access to information and resources (or adequate information ), and expressing ownership or feeling a valued team member (or ownership ). Also, there are high costs in terms of time, staffing, and other logistics associated with current models. Social media can promote connection both online and offline. It can als o provide quick access to a variety of information, resources, and knowledgeable others. Additionally, it can aid users in developing personal and professional identities and owning those identities publicly. It offers this all for simply the cost of opt ing in. Here, I elaborate on 37 these connections between socialization needs and affordances of social media, illustrating their potential synergies. Social media supports ownership. In social media, the participant drives participation by connecting to th e topics and people of interest, engaging at the desired level, and contributing as perceived as beneficial. So exposure and sharing can occur in ways that are relevant to individual needs. Though initial training and support may enhance the likelihood f or adjunct faculty members to initiate use of social media and to use it effectively, these tools usually have low learning curves and lend themselves to immediate benefits. Research has illustrated that social media can increase the volume and complexity of communications when utilized in a traditional learning environment (Brady et al., 2010; Schroeder & Greenbowe, 2009). Social media can enhance student -faculty connection, facilitating prompt feedback and information sharing along with collaborative lea rning and participant ownership of content and experience through active learning (Davis et al., 2012). It seems reasonable that these benefits could translate to the adjunct faculty audience in terms of their interactions with one another along with the full -time faculty and the administration. In this research, the concept of ownership is operationalized through items addressing factors such as oneÕs sense of being valued in work environment and serving as an integral contributor in it. Administration or departmental leads could capitalize on the needs expressed by adjuncts in this forum and plan responsively. For instance, face -to-face or synchronous opportunities for interactions and professional development might be planned around specific topics t hat adjuncts had initiated so that adjuncts would feel more heard, valued, and see the 38 immediate response to their needs or application of their ideas. Adjuncts could then quickly learn to leverage this social space in their practice and communication, fu rthering their investment in the community and in developmen t of oneself as a practitioner. Social media also allows for user -generated content and targeted communications to either individuals or to groups of users, allowing a platform for individual expression, reflection, feedback, and identity development. This is evident as academics are now capitalizing on the benefits of social media not only for personal connections but also for professional development and professional connections (Gruzd et al ., 2012; Moran et al. , 2011; Rowlands et al., 2011). Social media has been documented to aid the development and expression of identity (Grasmuch et al. , 2009; Greenhow & Robelia, 2009; Pempek et al. , 2009; Subrahmanyan et al. , 2008; West et al. , 2009). It provides a semi -public forum for sharing preferences, characteristics, interests, and affiliations in which peer feedback can help enrich relationships and support identity development (Davis et al., 2012). For first -generation students considering col lege, those who were connected on Facebook to someone that had enrolled in or graduated from college had higher confidence that they could attend and graduate from college (Wright, 1999). It seems this same logic could apply to work as a professional educ ator. Those observing successful faculty contacts online may feel more confident that they can succeed in instructional roles. In fact, the peer modeling that occurs on social media has also been linked to civic engagement (Beach & Doerr -Stevens, 2011; G reenhow, 2010; Robelia et al., 2011). Soc ial media supports access to information. As stated, social media has been shown to support complex communications (Brady et al., 2010; Schroeder & Greenbowe , 2009) by facilitating prompt information sharing and feedback (Davis et al., 2012). As 39 an online tool, social media is available at any time of the day in any geographic location that provides Internet access. It is even available on cellular data netw orks for individuals who own ÔsmartÕ or advanced mobile phones or digital devices. Once adjunct faculty begin connecting and communicating with each other via social media, they can assist each other by crowdsourcing to address questions, generate new kno wledge, troubleshoot problematic issues, or discuss practice. Wikipedia defines crowdsourcing as Òthe process of obtaining needed services, ideas, or content by soliciting contributions from a large group of people, and especially from an online community rather than from traditional employees or suppliersÓ (Crowdsourcing, n.d.). This takes the burden off of administration, relying more heavily on collective knowledge to address adjunct needs. Just as adjunct faculty members can target communications with one another based on discipline, need, or shared interest, administrators or mentors could connect with adjunct faculty members with customized information to assist them in meeting institutional, disciplinary, or role -related benchmarks. Connections and conversations can develop much more organically and democratically than in the prescribed, top -down structure prevalent in traditional online training and education environments, such as in learning management systems (LMS) like Blackboard, Moodle, and De sire2Learn. Though we often do not think of formalized onboarding initiatives occurring in constructivist ways, this style of learning or education can be quite conducive to formalized educational ventures. As discussed, several studies have illustrated that social media provides an awareness of diverse perspectives and opportunities for peer modeling, which can motivate behavior offline (Beach & Doerr -Stevens, 2011; Greenhow, 2010; Robelia et al., 2011). 40 Social media supports connectedness. Social medi a is largely intended for ongoing, asynchronous and synchronous participation. Platforms like Facebook, YouTube, and Twitter offer spaces for sharing images or videos, profiles detailing hobbies and interests, and connections to others. Some platforms, li ke Google Drive or Etherpad allow for collaborative editing, and tools such as Google+, even have methods for synchronous interactions with Web cameras and vocal dialogue. Ambient interactions with the user profiles and ÔstreamsÕ of others on social networ k sites have been shown to pave the way for future interactions by providing the impression that one is simply leveraging an existing connection (Greenhow & Burton, 2011; Steinfield, Ellison, & Lampe, 2008; Valenzuela et al., 2009). In one study, those en gaged in direct online communication activities with others, such as messages back and forth with contacts, were more likely to be engaged with others offline (Junco, 2012). Additionally, the integration of social media into course experiences has been linked to deeper interpersonal connections and positive psychosocial development for students, some of the reasons being that social media provides both an easy access point for connectedness through peer interactions and connectedness because it provides an opportunity for participants to receive social support from, and coordinate additional interactions with, their peers (events, text messaging, etc.) that transcend the online space (Heiberger & Harper, 2008; HERI, 2007; Junco et al., 2011). Academics have illustrated growing interest in connecting with others personally and professionally through social media (Gruzd et al., 2012; Moran et al. , 2011; Rowlands et al., 2011). Also, social media has been shown to support both bonding and bridging social capita l. Bonding social capital refers to resources or supports found within peopleÕs existing tight -knit groups 41 (close ties) and bridging social capital refers to resources or supports within groups of diverse others (loose ties) (Burke et al., 2010; Ellison et al., 2007; Ellison et al., 2014; Greenhow & Burton, 2011; Valenzuela et al. , 2009). This model also serves the small subset of users from this population that did not see connection with colleagues to be valuable, in that the interaction with others can be customized to more closely fit oneÕs needs. For instance, it may be that one has no desire to connect on a regular basis, but as a passive ÒnodeÓ in this available network, the individual can then opt in when connection is necessary for questions or co ncerns. Benefits of the new Moodle Socialwall format plug -in. Traditionally, social media outlets are public or semi -public, which has created anxiety about opting in for many educators and institutions in terms of maintaining a professional or institutio nal image and in terms of the Family Educational Rights and Privacy Act (FERPA) (20 U.S.C. ¤ 1232g; 34 CFR Part 99). This study leverages a newly constructed Moodle LMS format plug -in, the Socialwall, to test the concept of a natively -hosted social media community for adjunct socialization. In this approach, faculty need not have concern about public release of their comments and questions, as it is privately hosted on the institutionÕs servers with access available only for specified individuals. As def ined in Wikipedia (Plug -in (computing), n.d.), a plug -in is code that adds a specific feature or features to an existing software package, working in concert with that software as opposed to existing as a stand -alone software package. This format, when ap plied, alters the way a typical course looks and behaves to incorporate features and functionality typical to social media, while retaining the access to learning activities that are native within Moodle. 42 Specific to the constructs of socialization, Socia lwall features can provide support in a variety of ways. In terms of ownership, the Socialwall allows participants to ask questions and share on a wall or feed of activity, helping to drive and customize the interactions and content shared in the space. This allows for crowdsourcing and also allows administration to be both proactive and reactive in meeting the needs of adjuncts with comments, encouragement, and formal learning or interaction activities common to online experiences (polls, forums, surveys , learning modules, etc.). This may aid adjuncts in feeling that their needs are heard and addressed because they are a valuable part of the institution. The Socialwall also allows participants to construct a profile, sharing a picture of oneself along w ith contact information and other bits of share identity such as hobbies and interests. In this way, participants can practice publicly sharing and refining their professional identities. In terms of access to adequate information, I would again note that all participants are able to share on the wall. This sharing can include typed text, hyperlinks, and a variety of activities that are commonly available in LMS spaces (polls, forums, surveys, learning modules, etc.). Because the Socialwall exists online , participants are able to access information and contacts in the space at any time. In fact, they may even receive crowdsourced responses or interactions from peers outside of regular business hours. Participants will also have the option to engage in t he co -construction of knowledge with peers through activities such as dialogue on the Òwall,Ó interaction in forums, and hyperlinked objects and spaces such as Google documents, which allow for collaborative word processing and editing. This should aid ad juncts in gaining access to the information and resources that they need to succeed, along with allowing spaces for clarification of workplace and team expectations. In terms of 43 connectedness, participants are able to message collectively or individually to anyone existing in the space through the ÒwallÓ or through private messages in the ÒparticipantsÓ list. They will be able to share best practices, stories, words of encouragement, hobbies and interests, and more through these means. Seeing the posts o f others and the public selves those others present on their profiles in this space may ease introductions and continued communications for participants, supporting both bonding and bridging social capital. Participants will also be able to announce campu s and social activities that may generate further interactions face -to-face. In these ways, and in ways yet unanticipated, the features of the Socialwall have the potential to support the socialization of adjunct faculty. Conclusion. In the referenced mo del for socialization (Schrodt et al., 2003), a satisfying socialization experience may be represented by the summation of ownership, adequate information, and connectedness. This serves as the dependent variable for this study, which is represented by pa rticipantsÕ self -reported measures of socialization on a 12-point, three -construct scale. This research suggests opportunities for socialization can readily be facilitated at a low cost with low staff effort via social media technologies. This is due to the fact that social media participants are surrounded by information and knowledgeable people that provide access to adequate information, foster connectedness, and model and support ownership (see Figure 1 below). Also, I propose that an internally -host ed social media solution may be a beneficial choice for early experimentation to help assuage participant anxieties about participation. So, the independent variable is represented by assignment to an intervention group (or not) that makes use of the Mood le Socialwall format. Thus, in the hypotheses and research 44 questions, I aim to study the capacity of a simple group intervention deployed through the Moodle Socialwall format plug -in to impact the self -reported socialization scores of adjunct faculty at a small, Midwestern community college. Additionally, I explore the themes that emerge in intervention group communications as related to the constructs of socialization. Figure 1. Socialization via social media. This figure illustrates a conceptual frame in which the connected others and information in the social media environment aid a satisfying socialization experience by supporting oneÕs senses of ownership, connectedness, and access to adequate information. 45 CHAPTER III Methodology As introduced in chapter 1, this study aims to explore the potential of social media (i.e., the Moodle Socialwall format ) to facilitate positive socialization experiences for adjunct faculty. The overall goal of this work is to advance understanding of how to more effectively onboard adjunct faculty members so that they, in turn, might serve more effectively in their roles. To this aim, I posed the following research que stion along with several related hypotheses: RQ1: Do adjunct faculty participating in the Moodle Socialwall intervention report higher perceptions of socialization than their peers? Related to this research question, I proposed three hypotheses: H1: Adjunc t faculty participating in the Moodle Socialwall intervention will report higher average perceptions of socialization than their peers. H2: Adjunct faculty participating in the Moodle Socialwall intervention will report higher average perceptions of social ization following the intervention. H3: Adjunct faculty not participating in the Moodle Socialwall intervention will report consistent perceptions of socialization across time. Two additional research questions were posed to provide further context for th e results: RQ2: What themes in dialogue and participation emerged during the experimental group intervention? RQ3: What benefits, if any, did the experimental group perceive that the intervention provided to support their socialization? 46 In relation to these questions, I predict that emerging themes from intervention interactions will be related to the three constructs of socialization (ownership, adequate information, and connectedness) and that, in general, the experimental group will perceive that the intervention supported their socialization. To address these research questions, this study design most resembles the pre -test and post -test control -group design presented by Creswell (2014) (see Figure 2.1). This design was selected with guidance from a dissertation committee member who is a practiced quantitative researcher as one that would more immediately aid me in addressing my hypotheses. Figure 2.1 illustrates how participants were randomly assigned to either an experimental or a control group an d only the experimental group had access to the intervention, though both groups received a pre - and post -test. This design was ideal to explore the efficacy of the new Moodle LMS Socialwall format plug -in as a low -cost solution for the socialization of a djunct faculty members because it would allow one group to experience potential benefits while holding the other group in somewhat of a constant state. In addition, this study design leverages mixed methods to provide additional context for results and th emes. Collecting qualitative data aided in exploring how a specific population of adjunct faculty members perceived that participation in the intervention impacted their socialization experience at the institution. Throughout the study, the independent v ariable was assigned to either the experimental or control group and the dependent variable was participantsÕ self -report of socialization measures. In the following sections, I will provide more details on the sample, the instrumentation, the procedures, and limitations of these methods. 47 Group A R Ñ O ÑÑ X ÑÑ O Group B R Ñ O ÑÑÑÑÑ O Figure 2. Nonequivalent (pretest and posttest) control -group design. Rationale for Research Approach I began this proposal envisioning a phenomenological study to understand the experience that adjunct faculty might have in a socialization intervention leveraging social media. However, after much consideration coupled with the guidance of my dissertation committee, I acknowledged a personal desire to focus on exploring the efficacy of social media in enhancing the socialization experience for adjunct faculty. This led to the current pre -test post -test control group design. An action research approach wa s also considered, and the reflection and community aspects of action research could be incredibly useful in designing programming to meet adjunct faculty socialization needs. Yet, the design of a program spans beyond the scope this dissertation study. Q uantitative research is somewhat limited in its capacity to tell stories or provide for consideration of the rich context of a response. For these reasons, I have included some qualitative data collection to complement my quantitative methods. Research Sa mple A population of 249 adjunct faculty members from a small, Midwestern community college were purposively recruited for participation in the research via email. Previous research indicated that this population was 52% male, 48% female, 88% Caucasian, and held a mean age of 46 with modes at 29 and 49, and an approximate 7.8 years of service as adjuncts at the surveyed institution (Wicks, 2013). From this adjunct population, 43.37% (N=108) consented to the research. Out of these consenting participants , 90 completed the initial pre -test (36.14% of the population or 83.33% of 48 those consenting to the research). Then, from that subset, 77 completed the post -test, which equates to 30.92% of the population. Of the final sample of 77 participants, 42.9% were male (N=33) and 57.1% were female (N=44). The mean age was 45.3 with modes at 30 and 32 (and a range of 23 to 72 years of age), and they had attained an average of 6.66 years of service (and a range of less than one year to 33 years of service). So, in comparison to the demographics of the larger adjunct population at this institution, the sample of 77 was a bit younger, with a bit less teaching experience at the institution, and notably more skewed toward female participants as compared to male particip ants. All participants were active adjuncts at the surveyed institution; 96.1% of the sample was (N=74) currently teaching for the institution at the time of the intervention. Of those, 18.2% (N=14) were teaching online only and 19.5% (N=15) were teachin g only at off -site extension locations for a total of 29 participants (37.7%) that did not have regular access to the physical college campus. In addition to this, it is important to note that 22% (N=17) of adjunct faculty participants also held dual stat us in administration or the hourly support staff team at the surveyed institution. Sampling design. To ensure ethical protection of human participant rights, all proposed methods were run through an institutional review board at Michigan State University and gained approval. The study utilized an inclusive sampling design, which invited all 249 adjunct faculty members currently on the adjunct faculty email listserv at the institution to participate. The automated adjunct faculty email listserv includes a ny active adjunct faculty, with active being defined as any adjunct faculty that have been assigned in the student information system as the instructor of a class being offered 49 through the institution within the past year. Any adjunct who has not been ass igned as the instructor of a class within the previous calendar year is considered inactive and would not have been included in this call for participation. Power analysis. An a priori power analysis for a two -group independent sample t-test was performed with the G*Power 3.1 software package to approximate the ideal number of participants for detecting potential results from the proposed intervention. The control group mean (29.26) an d standard deviation (9.57) were drawn from responses in a previous study utilizing the same instrument with the same population (Wicks, 2013) while the values for the experimental group were drawn from a pilot study of these specific research tools with t he target population (M=26.55, SD=6.27). Using this information, G*Power indicated an effect size of .336. At a power of .80, G*Power suggested that the ideal sample size might be approximately 222 participants with 111 in each group and a critical t val ue of 1.65. With a population size of 249, this seemed quite optimistic, thus I employed a compensation model for participation in both of the surveys (pre-test and post -test) and in the intervention. The compensation consisted of $5 per completed survey and an opportunity for intervention participants to enter gift card drawings to promote sustained participation. Instrumentation The instrumentation consisted of a pre -test (see Appendix E), a post -test (see Appendix I), along with archived activities wit hin the intervention space (i.e., the Socialwall in the Moodle LMS). Both the pre -test and post -test were hosted by and deployed through the Qualtrics online survey platform. These tests contained Likert -type questions related to participantsÕ socializat ion and colleague interactions. For instance, 50 this scale contained items such as ÒI feel connected to others in my work environmentÓ and ÒI have been given adequate information on how to serve the institution effectively.Ó Participants were asked to resp ond on a 5 -point scale ranging from Strongly Agree to Strongly Disagree. The pre -test also contained a set of Likert -type questions related to themes that had emerged in prior research with the population. Examples here contained items such as ÒAdjunct i nput is valuedÓ or ÒI would participate in evening opportunities for involvementÓ with the same 5 -point rating scale. Furthermore, the post -test contained a set of Likert -type questions related to the experimental group intervention. These items included prompts such as ÒI look forward to future opportunities to use the SocialwallÓ or ÒParticipation exposed me to information that is relevant to my service as an adjunct.Ó In addition, both instruments collected responses to open -ended questions on socialization, colleague interactions, and social media use along with general demographics. Additional questions such as those about social media use and the frequency and nature of interactions with colleagues that were asked on both the pre -test and po st-test were useful in providing more context to gauge change in those areas as potential themes. The post -test posed additional open -ended questions related to the experimental group intervention. For instance, in order to gather data on whether and how t he experimental group perceived the intervention supported their socialization (RQ 3), the post -test included open -ended questions about what features of the Socialwall were mo st important in supporting participantsÕ sense of ownership for their primary ac ademic departments, what features were most important in supporting access to relevant 51 information, and what features were most important in supporting their sense of connectedness to their colleagues. The intervention itself was also used as a source of d ata, given that the LMS environment allows for collection of data related to user activity, including user comments and responses. Participation statistics and responses from all Moodle Socialwall activities, including the Socialwall feed along with comme nts from other activities used in the space like forums and surveys, were collected to help address what themes in participation emerged (RQ 2). Scale Reliability Schrodt et al. (2003) utilized 16 items from Cawyer & FriedrichÕs (1998) previous research on faculty socialization. As Schrodt et al. explain, ÒAlthough Cawyer and Friedrich (1998) demonstrated construct and face validity for the socialization items, they did not test the underlying factor structure to determine how many dimensions of socializat ion were reflected in the 16 -item scale.Ó (p. 22). Schrodt et al. performed a principal components factor analysis to address this, which evidenced a three -factor solution that accounted for 65.20% of total variance: ownership ( ! =.87), adequate informatio n (! =.83), and connectedness ( ! =.84). Only 14 of the 16 items corresponded to one of these three factors, so the other items were dropped in analysis. According to their research, this scale for socialization has strong internal consistency. I adapted a version of this scale in previous research with the adjunct population at the surveyed institution. In this adaptation, two questions from Schrodt et al. (2003) were dropped due to advice to limit the length and repetitiveness of the survey tool, which eq ualized the number of questions per construct to four and to provide a total of 12 52 questions. In addition, very slight changes were made to the wording of questions to use language that was more suited toward the surveyed population and their work environ ment. In that study, CronbachÕs Alpha scores were quite high, ranging from .803 to .842 on the three constructs and .903 on the scale overall. In the current study, I again used the adapted scale for socialization inspired by the work of Cawyer & Friedri ch (1998) and Schrodt et al. (2003). Pallant (2013) reports that any CronbachÕs Alpha values above .7 are generally acceptable, though values of .8 and above are preferred, which suggest ed very good internal consistency and reliability for the scale with this sample based on that previous population research . Procedures In a previous pilot study conducted in 2013, in which many of the adjunct faculty population had participated (Wicks, 2013), I informed participants that I would solicit their participation in future research. In January, I spread the word about my dissertation study to the adjunct population at events like the faculty orientation and through individual conversations. Specifically, I told them I would be recruiting for research participants in one to two months. During February, I worked with the LMS administration team at our institution to acquire the software plug -in and install and test it on our system. I will discuss the software itself along with details of how the research process progressed in the paragraphs that follow. The Moodle Socialwall format plug -in. The Socialwall format plug -in was commissioned by Chris Kenniberg and Bryan Smith on the website moodle.org (Moodle Facebook/Social Course Format, 2014). I referenced a blog post on this topic while searching Moodle.org for existing social media -related plug -ins. Moodle is an open 53 source LMS, which means that users are able to add to its core functionality as they see fit. A community of developers has submitted their additi ons to the Moodle.org site for ease of sharing and cataloging. After reading the post, I contacted Chris Kenniberg by email to learn more. Chris and I are both on the planning committee for the Michigan Moodle Moot, which is a statewide conference for us ers of this LMS. When Chris and I eventually spoke via telephone about the project, he explained that Synergy Learning, a Moodle Partner from the United Kingdom, had offered to develop this plug -in. Also, he indicated that the REMC Association of Michiga n, one of the co -sponsors of the Michigan Moodle Moot, had agreed to fund the development. This seemed a perfect tool for my research topic because it provided the natively -hosted social media option I sought along with basic features of social media func tionality. Once this software was complete for an initial release to the Moodle community, Chris shared the code, the programming team at the surveyed Midwest community college installed it, and the programmers and I pilot tested the software to ensure th at it was working as anticipated. Specific to the constructs of socialization, Socialwall features can provide support in a variety of ways. In terms of ownership, the Socialwall allows participants to ask questions and share on a wall or feed of activity (see Figure 2), helping to drive and customize the interactions and content shared in the space. This allows for crowdsourcing and also allows administration to be both proactive and reactive in meeting the needs of adjuncts with comments, encouragement, and formal learning or interaction activities common to online experiences (polls, forums, surveys, learning modules, etc.). This may aid adjuncts in feeling that their needs are heard and addressed because they are a valuable part of the institution. T he Socialwall also allows 54 participants to construct a profile, sharing a picture of oneself along with contact information and other bits of their identity , such as hobbies and interests. In this way, participants can practice publicly sharing and refinin g their professional identities. In terms of access to adequate information, I would again note that all participants are able to share on the wall. This sharing can include typed text, hyperlinks, and a variety of activities that are commonly available i n LMS spaces (polls, forums, surveys, learning modules, etc.). Because the Socialwall exists online, participants are able to access information and contacts in the space at any time. In fact, they may even receive crowdsourced responses or interactions from peers outside of regular business hours. Participants will also have the option to engage in the co -construction of knowledge with peers through activities such as dialogue on the Òwall,Ó interaction in forums, and hyperlinked objects and spaces such as Google documents, which allow for collaborative word processing and editing. This should aid adjuncts in gaining access to the information and resources that they need to succeed, along with allowing spaces for clarification of workplace and team expe ctations. In terms of connectedness, participants are able to message collectively or individually to anyone existing in the space through the ÒwallÓ or through private messages in the ÒparticipantsÓ list. They will be able to share best practices, stori es, words of encouragement, hobbies and interests, and more through these means. Seeing the posts of others and the public selves their colleagues present on their profiles in this space may ease introductions and continued communications for participants , supporting both bonding and bridging social capital. Participants will also be able to announce campus and social activities that may generate further interactions face -to-face. In these ways, and in ways yet unanticipated, the 55 features of the Socialwa ll have the potential to support the socialization of adjunct faculty. Figure 3. The Moodle Socialwall. This figure depicts a sample installation of the Moodle Socialwall format plug -in. 56 Intervention prompts. To encourage interaction in the space, and to further explore themes related to the constructs of socialization, eight intervention prompts were authored (see Appendix K). The prompts were to be posted in the space at a rate of 2 per week for each of th e 4 weeks of the intervention. One of the prompts each week would serve to promote participant interaction while the other prompt for the week would serve as both a conversation starter and as an activity that participants would need to complete in order to be eligible for a $100 gift card drawing . In this way, the drawing prompts also served as incentives for continuing participation. An example of a prompt designed to explore the construct of ownership is as follows: Your continued service is key in m eeting student demand. Does anyone want to guess how many adjuncts are in our active pool at MMCC? What about the current adjunct to full -time ratio? What impacts, if any, do you think this staffing model has on your teaching, your course design, or on general strategic directions in which our academic departments are headed? By stating that their service is key in meeting demand, I am suggesting that they have value to the institution. Though the ratio is just a bit of trivia, it also provides further information for these adjuncts on the make -up of institutional staffing and what a large role they play in that. Then, by asking what impacts this staffing model may have on their teaching or design or strategic directions of the department, I am asking them to talk more about their ownership of the role in terms of their individual teaching activities and in relation to the role they play with the academic departments. Here is a sample of a prompt designed as both a drawing requirement and a tool to expl ore the construct of adequate information: IÕve started a Moodle Choice Poll below. If you opt into this poll by selecting a response by this Thursday, March 12 at 12PM EST, you will be entered into a drawing for a $100 gift card. Out of those who respond by 57 that date, a winner will be selected through the aid of a random number generator. Poll: Do you intend to participate in our faculty professional development day on Wednesday, April 1, 2015? For the fir st time, four different break -out sessions for adjunct faculty will be available in the afternoon while full -time faculty attend Faculty Senate. For those of you that havenÕt previously attended a PD Day, there is compensation available for your participa tion. A full schedule for this event will be sent out by Academic Council in the near future. By stating that I am utilizing the Moodle Choice Poll Activity, I am providing more information to them on the activities innate to their LMS. I am also providing in -depth information about processes that will be used to select drawing winners, so that th ey feel informed. Then, I introduce the upcoming faculty professional development day, which is an event centered around providing all faculty with access to information and resources that will aid them in their practice. I am also explaining how the bre ak-out sessions will be specifically relevant to them as the audience. Pilot of survey instruments and prompts. During this January to February 2014 timeframe, I also piloted initial versions of the pre -test and post -test surveys. The pre-test and post -test surveys were sent by email through Qualtrics to eight members of the target population that would not be participating in the actual research and to three of the faculty serving as doctoral dissertation committee members for this project. The interven tion prompts were also shared with several population members for feedback. I followed up with these individuals through email and in person to ask about survey features such as phrasing of questions, question order, length, and so forth. Adjustments were made accordingly. Design of survey communications (pre -test and post -test). The products and approach involved in participant communications were heavily influenced by Dillman, 58 Smyth, & ChristianÕs (2009) tailored design method for mail and internet sur veys. In DillmanÕs schematic overview of the tailored design perspective (p. 27), he provides tactics to establish trust, to increase social rewards, and to reduce social costs. Many of these tactics, such as sponsorship by legitimate authority, tangible rewards, support of group values, and utilizing short questionnaires, were employed in my methods. He also suggests that a well -formulated survey process will include at least four structured communications that employ these tactics. My communications i ncluded approximately five structured communications that were sent to all participants, including the invitation to the research and consent form, the group assignment and pre -test, compensation for completing the pre -test, the thank you and post -test, an d the compensation for completing the post -test. In addition to these communications for all participants, those that did not respond promptly to the consent, pre -test, and post -test were sent reminder communications. All communications, survey tools, an d communication management was initiated through Qualtrics survey software. Compensation structure. Research shows that reasonable financial incentives have been widely documented to increase response rates in surveys and other research activities (Church , 1993; Dillman et al., 2009; Singer, Van Hoewyk, Gebler, Raghunathan, & McGonagle, 1999). For these reasons, coupled with the high number of ideal participants preferred by the power analysis, I e mployed a compensation strategy that was laid out to parti cipants in the informed consent form (see Appendix B). For instance, I offered a $5 incentive for each of the two surveys completed per participant. A second part of the strategy related just to the experimental group while engaged in the intervention. It was the drawing of four, $100 gift cards to activ e participants throughout 59 the one -month intervention. The goal of these drawings was to encourage and sustain participation across the month that data was being collected. The process for awarding these compensations will be described in further detail w hen I discuss the procedures related to the intervention. Recruitment and group assignment. An initial, two -week call for participation was made via email (see Appendix A) to the full adjunct faculty population listserv at the institution through Qualtric s. This email included a link to a consent form (see Appendix B) created with the survey tool in Qualtrics. That consent form contained standard information about the research purpose, details, and contacts along with a single question regarding their co nsent to participate. After one week, a reminder email was sent to the population members that had not yet consented through Qualtrics. It again included a link to the consent form. Once the consent survey closed in Qualtrics, the spreadsheet of respond ents was downloaded. With the sheet ordered according to time of response, every other participant on that sheet was assigned to the experimental group. This approach was used to negate any potential effects related to having more Òearly respondersÓ or Ò late respondersÓ in either of the groups. Participants were notified of their group assignment in the next communication, which prompted them to complete the pre-test. Pre-test and compensation. In the next email, participants were informed of their assi gnment to either the control group (see Appendix C) or the experimental group (see Appendix D). This email was sent via Qualtrics only to the population members that had consented to participate in the study. The email also included a request for partici pants to complete the pre -test in Qualtrics within the space of one week. For the 60 experimental group participants, the note also provided instructions on how to access the intervention space in Moodle once they had completed the pre -test. After four days , reminder emails (see Appendices F & G ) were sent to the participants that had not yet completed the pre -test and also to those that had not yet accessed the intervention space. All participants that completed the pre -test, aside from several that had op ted out of compensation, received a $5 bill with a note signed by me that thanked them for their assistance through interoffice or postal mail. After one week had passed and a reminder had been sent, any experimental group participants were removed that h ad not both completed the pre -test and accessed the intervention space. Moodle Socialwall intervention. During the course of one month of interactions in the Moodle Socialwall intervention space, participants were instructed to use the space as they saw fit and interact at a level that was useful to them as they would in a traditional social media space. In addition to the compensation of survey participants to increase response rate, the design provided compensation for intervention participants because research has shown that incentive mechanisms can also be successful in increasing contributions to online learning networks (Hummel et al., 2005). Thus, in efforts to build interest and sustain engagement in the intervention, I created the prompts previo usly discussed, including activities related to the four gift card drawings offered throughout the duration of the experiment (see Appendix K). This also aided in providing a baseline level of interaction in the space, which is necessary in the Schrodt e t al. (2003) model. The intervention prompts were posted at a rate of two each week. The drawings were available to active participants, as defined by those participants having completed 61 an LMS prompt activity that was indicated as required to opt into th e drawing . Those not responding to the prompt activity were not eligible for the weekÕs drawing even if they were otherwise active witihin the Socialwall space. These necessary activities included: responding to a poll about attendance at an upcoming eve nt, responding to a two - question survey on professional development interests, posting in a forum related to educational technology and teaching needs, and creating a glossary entry on a hobby or interest. At the point that the drawing entrance deadline passed, all participants that had completed the tasks necessary to be included were put in a numbered list in chronological order of their participation. Then, a random number generator was used to select the winner of the drawing . The winner of each drawing immediately received a $100 Amazon.com giftcard delivered via email and the experimental group was informed of the winner through the News forum in the course space, which also emails out to participants. Non -drawing prompts included activities such as guessing facts about the adjunct faculty body, providing input on the Socialwall tool and how it functioned, participating in forums related to adjunctsÕ work at the College, and providing input on current socialization efforts. Post -test and compensati on. In the final email (see Appendix H), participants were thanked for their efforts and asked to complete a post -test (see Appendix I) in Qualtrics within one week. A reminder email (see Appendix J) was sent after four days to the participants that had not yet completed the post -test. All participants that completed the post -test, aside from several that had opted out of compensation, received a $5 bill with a note signed by me that thanked them for their assistance through 62 interoffice or postal mail. After the post -test deadline had passed, all participants that had not completed the post -test were removed from the study. Data Analysis Methods As previously stated, prior to any comparison of values related to the hypotheses on socialization, CronbachÕs Alpha scores were computed for the socialization scale to ensure internal consistency and reliability. Additionally, correlations were investigated with Pearson product -moment correlations between socialization scores on the pre -test and population demogr aphics such as age, gender, years of teaching experience at the institution, and off -campus status. This aided in determining if additional group means should be compared to inform results. Data analysis for hypotheses. Once these preliminary calculation s were complete, attention was turned to the first research question ( Do adjunct faculty participating in the Moodle Socialwall intervention report higher perceptions of socialization than their peers? ) and accompanying hypothesis: H1: Adjunct faculty participating in the Moodle Socialwall intervention will report higher average perceptions of their socialization than their peers. Post -test means from both the control and experimental groups were compared with an independent samples t -test in SPSS Version 23. The assumptions that typically apply to parametric tests, like t -tests, include: measurement of the dependent variable on a continuous scale, random sampling, independent observations, normal population distribution, and homogeneity of variance (Palla nt, 2013). While the first three assumptions noted here were accurate, I did note slight differences between the sample and the population characteristics as represented in a previous study (Wicks, 2013). This 63 sample was younger, with a bit less teaching experience at the institution, and notably more skewed toward female participants than male participants as compared to the overall target population. In terms of homogeneity of variance, the LeveneÕs test utilized during the independent samples t -test i ndicated significance values higher than .05, which suggests that the variability of scores for each group is similar (Pallant, 2013). Additionally, the participants that held dual status as administrators or hourly support staff members (22% or N=17) wer e removed from the data set and the t -test was run again to examine results for only those holding adjunct status. These groups were explored separately because those holding dual roles at the institution often have access to more opportunities for formal training and interactions with colleagues than do those holding only the adjunct role, which had the potential to impact results. The second and third hypotheses were as follows: H2: Adjunct faculty participating in the Moodle Socialwall intervention w ill report higher average perceptions of their socialization following the intervention. H3: Adjunct faculty not participating in the Moodle Socialwall intervention will report consistent perceptions of socialization across time. Because these hypotheses r equired comparison of a single groupÕs scores from pre-test to post -test, paired samples t -tests were used. Again, after the initial results were run, those holding dual status as administrators or hourly staff members were removed from the data set and c alculations were run again. The means of the control group on pre- and post -tests were compared across time through a paired samples t -test. This same 64 procedure was used to compare means of the experimental group from pre -test to post -test. Data analysis for research questions. Additionally, I posed two research questions, as follows: RQ2: What themes in dialogue and participation emerged during the experimental group intervention? RQ3: What benefits, if any, did the experimental group perceive that the intervention provided to support their socialization? The second research question ( What themes in participation emerged during the experimental group intervention?) was analyzed via thematic content analysis (Glesne, 2011). A coding template that summari zed and organized major themes in the data from the Moodle Socialwall LMS space was developed; with counts being added each time those themes appeared (see Appendix L). This technique was used as opposed to other forms of analysis in qualitative research such as conversion analysis, narrative analysis, or grounded theory (Glesne, 2011), because of the concrete aims of the question. Strengths of thematic analysis are that it allows for flexiblilty in terms of the applications of multiple theories and also that it allows for categories to emerge from the data (Thematic analysis, n.d.). However, there are also challenges associated with this technique such as the concern for the reliability of results and the potential of missing nuanced data (Thematic analy sis, n.d.). The goal of the question was to add context to this largely quantitative report on the potential of the Socialwall tool in terms of bringing the vibrancy of participant comments in to illustrate scope and impact. 65 There are six main steps or phases common to thematic analysis: (1) becoming familiar with the data, (2) generating initial codes, (3) searching for themes, (4) reviewing themes (5) defining and naming themes, and (6) producing the report (Thematic analysis, n.d.). Though I acknowle dge that the model of socialization utilized in this research provided a reference point as I analyzed this data, I wanted to take a more grounded approach to developing codes and themes. Though I was not engaging in grounded theory research, which builds theory by searching for themes and patterns in the data (Glesne, 2011), I used the patterns emerging in the data as a starting point for establishing codes and themes (see Appendix L). During phase 1, I exported the 144 written posts, comments, and respo nses from participants in all activities that were part of the Socialwall intervention experience into a word processing format. During phases 2 and 3, I assigned and refined 41 initial, rudimentary codes to these data points by hand based on my own initi al perceptions refined in alignment with emerging themes. For instance, in preliminary coding a variety of codes related to technology were present (LMS, Moodle, the Socialwall, technical support, & technology) which were later refined just to technology to represent the larger theme. Given that RQ2 did not specifically ask about themes related to socialization, I did not initially attempt to phrase the data through that lens. In phases 4 and 5, I further refined these themes, making note of their frequen cies by hand, and attempted to relate the 5 emerging themes to the three constructs of socialization discussed in this study. In taking the example of the technology theme above, here I expanded the theme to discuss the specific sub -themes under that item , such as desire for technology resources, support in the use of technology, and institutional use of technology to connect with the adjunct audience. 66 The third research question (How, if at all, did the experimental group perceive that the intervention su pported their socialization?) was explored in three ways. These first included trends in Likert -type scores regarding the intervention experience and its support for each of the three constructs of socialization, along with thematic content analysis of co mments and posts made in the intervention space about the Moodle Socialwall format plug -in and, lastly, thematic content analysis of open -ended survey responses from intervention participants on the post -test. The Likert -type scores were compared in terms of mean scores with standard deviations noted along with percentages for responses at the low, mid, and high point of the scales. The 75 open -ended survey question responses were exported collectively into a word processing document and then coded with t he thematic content analysis procedure previously described in detail. Initial codes such as accessibility, collaboration, and none were created. Then, these codes were refined and further examined to identify 5 emerging themes containing multiple codes. For instance, the theme of Connection contained codes like collaboration, connection, help, and sharing viewpoints. Lastly, the Socialwall comments exported for research question two were reviewed to extract all Socialwall comments related to the Moodle Socialwall itself. In those 26 posts, an abbreviated thematic content analysis process was performed to identify 6 themes in these responses. Reliability and Validity Internal validity. First, I will explore a variety of threats to internal validity. Internal validity refers to ÒÉexperimental procedures, treatments, or experiences of the participants that threaten the researcherÕs ability to draw correct inferences from the data about the populationÉÓ (Creswell, 2014, pp. 174 -175). Threats related to ti me include 67 history and maturation , or the threat of external events or participant growth that influence the outcomes beyond the experimental treatment, therefore influencing results (Creswell, 2014). To protect against these risks, I executed the research over a fairly brief one-month timeframe with both an experimental and a control group comprised of adults in similar working conditions with similar daily routines that were surveyed concurrently. Risks related to sampling include both regression , or par ticipants with extreme scores that regress across time, and selection , or participants with certain characteristics that predispose specific outcomes (Creswell, 2014). To mediate the potential impact of extreme scores or participant predispositions, I sea rched for outliers in demographics and responses and utilized an open call for participation with the population that included a randomized assignment to one of two groups. In terms of the tools, effects of testing (participants memorizing tools to fill i n similar responses later) and instrumentation (changes in instruments that impact scores) are both risks (Creswell, 2014). To lessen these impacts, questions related to the socialization scale were re -ordered from pre -test to post -test, though the survey item itself remained consistent. One must also consider threats related to compensation and the potential benefits of the intervention, which might include compensatory or resentful demoralization and compensatory rivalry , or the risks that one group wil l feel resentment or impact effects if the other receives benefits perceived as more desirable (Creswell, 2014). To assist in controlling for these, all participants were told that they would have access to test the tool and interact following the brief, one-month intervention and both groups were provided with some level of compensation for their participation. Finally, in terms of participant behavior, mortality , or the loss of participants, and diffusion of treatment , or contamination of the control 68 group by the treatment, are both risks (Creswell, 2014). To mediate these risks, I provided compensation opportunities throughout the experience to maintain participation and I deployed the intervention asynchronously online with a geographically dispersed audience, which aided in keeping the groups somewhat separate. External validity. In contrast to internal validity, which focuses on methodology, external validity refers to the degree to which results are generalizable to other populations, settings, and time periods (Creswell, 2014). There are also several threats to external validity posed by Creswell (2014, p. 176). Interaction of selection and treatment is the idea that the results cannot be generalized outside of the specific population characteristics while interaction of setting and treatment is the idea that results cannot be generalized outside of the specific setting characteristics. Lastly, inter action of history and treatment is the idea that results cannot be generalized to past or future situations. I address these threats by recognizing that there is no generalizability of th ese results outside of this unique context. The individuals and cul ture at this small college are quite unique given a variety of factors such as its geographic location in proximity to other institutions, the socioeconomic factors in this area, long -standing political perspectives at the institution, and financial constr aints and funding models among a host of other things. Likewise, this historical time period is very unique for the institution given that it has both seen its student population double in the last 10 years and then decline somewhat in recent years. Thus , the results of this study are not meant to apply to other groups in other locations, but simply to add to our understanding of the dynamics at this location and to provide a model that may have merit in trial elsewhere. 69 Qualitative reliability and validi ty. Though the use of qualitative data was minor, I also employed several strategies outlined by Bashir, Afzal, & Azeem (2008) to promote reliability and validity with my qualitative research techniques. First, I leveraged multi -method strategies to allo w for triangulation in data collection and analysis. Data was collected through numerical scales, open -ended questions, participant intervention comments, and through quantitative data related to user activity in the Socialwall LMS space during the interv ention. Additionally, the survey scale is drawn from, or inspired by, an existing scale model referenced in the area of faculty socialization that has evidenced high internal reliability and consistency. Second, the inclusion of open -ended survey questio ns and intervention participant comments will also allow me to share verbatim accounts from participants on their experiences. The inclusion of literal statements and participant quotations should allow readers to better evaluate my claims. Third, in a s ense, I have the benefit of multiple researchers. Given that I am a doctoral candidate and this research is being proposed and executed to meet the dissertation requirement of that experience, I work closely with a primary advisor in crafting my research design and have the benefit of an extended guidance committee as well. All of these individuals are seasoned and respected researchers with experience in qualitative methods. Fourth, by deploying the intervention within Qualtrics and an LMS, I have the b enefit of mechanical data. This can be reviewed and recoded as necessary and leveraged in ways consistent with Institutional Review Board approval to corroborate the findings. Fifth, I have considered discrepant data, actively analyzing and reporting any data that are an exception to identified trends. It is my hope that the combination of these various techniques will provide a level of credibility, transferability, and 70 dependability to this research. Finally, I have acknowledged my bias in the introdu ction to this study by explaining my background with technologies and my inclination toward self-driven learning with these tools. 71 CHAPTER IV Results This study investigated 3 research questions: RQ1: Do adjunct faculty participating in the Moodle Socialwall intervention report higher perceptions of socialization than their peers? RQ2: What themes in dialogue and participation emerged during the experimental group intervention? RQ3: What benefits, if any, did the experimental group perceive that the intervention provided to support their socialization? In the paragraphs that follow, I present the results for the first research question along with its three hypotheses. Related to these hypotheses, data from a variety of t -tests is reported. Next, I present results for the second research question, including major themes of interactions as illustrated through participant comments and also LMS use statistics. Lastly, I present the results of the third research question, providing analysis of comments both from the intervention and the post -test. I close with a summary of the major findings, which will lead into the discussion of these results in the following chapter. Research Question 1 The first research question was Do adjunct faculty participating in the Moodle Socialwall intervention report higher perceptions of socialization than their peers? Three hypotheses were formulated related to this research question, as discussed below. This question was primarily analyzed utilizing t -tests to compare group means. In the next several paragraphs, I discuss the results of these tests. 72 Hypothesis 1. The first hypothesis proposed that Adjunct faculty participating in the Moodle Socialwall intervention will report higher average perceptions of their social ization than their peers. Once my initial analysis of the quantitative survey data was complete, I focused my analysis to address the first hypothesis with the aid of SPSS Version 23. This primary hypothesis was analyzed using an independent -samples t -test with the significance level set at .05 to analyze group mean composite scores for socialization. LeveneÕs test for equality of variances suggested that equal variances could be assumed (p=.22). There was no significant difference in scores when compar ing means of the control group (M=41.22, SD=9.35) to the experimental group (M=43.71, SD=8.01); t (75) =1.21, p= .23, two -tailed). The post -test group means for the three individual constructs of socialization (ownership, adequate information, and connecte dness) were also compared with independent samples -t-tests. Like the overall composite, the construct composites yielded no significant results. Additionally, a post hoc power analysis performed with the G*Power 3.1 suggested that the power of this test was approximately .23, which illustrated that the small sample size may impact the significance of results. Then, participants with affiliations to other employee groups such as the administration and hourly support staff were removed from calculations to compare Òonly -adjunctÓ groups. Again, these groups were explored individually because those holding other part - and full -time roles at the college often have additional opportunities for formalized training and interaction with colleagues as part of their work roles that those serving only in the adjunct role do not have. That difference in opportunities had the potential to impact results. Here again, there was no significant difference in mean 73 scores when comparing the post -test means of the control gro up (M=39.82, SD=8.73) to those of the experimental group (M=43.46, SD=8.05; t (58) = -1.66, p= .10, two -tailed) , though the general trend was that the experimental group had higher means . Therefore, hypothesis 1 was rejected. Hypothesis 2. H2: The experimental group of adjunct faculty will report higher average perceptions of their socialization on the post -test than on the pre -test. The second and third hypotheses required a different type of t -test for the comparison of the mean scores of a singl e group at different points in time. Again, analysis was facilitated with SPSS and a significance level set at .05. The second hypothesis, a predicted increase in experimental group composite socialization score means from pre - to post -test, was examined with paired samples t -tests (see Table 1). There was not a statistically significant change in mean scores from pre -test (M=44.32, SD=8.46) to post -test (M=43.71, SD=8.01), t (30) = -.71, p= .48 (two -tailed) here either. The mean change in socialization scores was .61 with a 95% confidence interval ranging from -1.15 to 2.37. Individual construct composites for the experimental group were also compared. No significant results were illustrated in those tests. Those holding dual status as administrators or hourly staff members were again removed from the data set and an additional paired samples t -test was run for the adjuncts -only group, yielding no significant results (see Table 2 for means) . Therefore, hypothesis 2 was rejected. 74 Table 1 Paired -samples t -tests comparing within -group pre- and post -test composite means Means Compared Group Significance (2 -tailed) Socialization Experimental .483 Socialization Experimental (Only Adjunct) .494 Socialization Control .061 Socialization Control (Only Adjunct) .048* Ownership Experimental .553 Ownership Experimental (Only Adjunct) .802 Ownership Control .045* Ownership Control (Only Adjunct) .061 Adequate Information Experimental .936 Adequate Information Experimental (Only Adjunct) .797 Adequate Information Control .052 Adequate Information Control (Only Adjunct) .044* Connectedness Experimental .374 Connectedness Experimental (Only Adjunct) .282 Connectedness Control .764 Connectedness Control (Only Adjunct) .563 *Significant at the .05 level Hypothesis 3. H3: The control group of adjunct faculty will report consistent perceptions of their socialization on both pre -test and post -test. The third hypothesis, a prediction of mean score stability for the control gro up socialization composite across time, was tested using paired samples t -tests (see Table 1). Again, analysis was facilitated with SPSS and a significance level set at .05. There was not a statistically significant change in mean scores from pre -test ( M=42.70, SD=8.99) to post -test (M=41.22, SD=9.35), t (45) =1.92, p= .06 (two -tailed). The mean change in socialization scores was 1.48 with a 95% confidence interval ranging from -.07 to 3.03. In testing the construct composites, there was a noteworthy re sult related to the control group mean scores for ownership (see Table 1). A paired -samples t -test identified a statistically significant decrease in control group ownership mean scores from pre -test (M=14.91, SD=3.57) to post -test (M=14.26, SD=4.11), t ( 45) =2.07, p=.05 (two -tailed). 75 The mean change in ownership composite scores was .65 with a 95% confidence interval ranging from .02 to 1.29. The eta squared statistic (.09) indicated a moderate effect size. Again, after the initial results were run, tho se holding dual status as administrators or hourly staff members were removed from the data set. With this adjunct -only group, a paired samples t -test (see Table 1) revealed that there was a statistically significant decrease in socialization means from p re-test (M=41.76, SD=8.65) to post -test (M=39.82, SD=8.73), t (33) =2.06, p= .048 (two -tailed). The mean change in socialization scores was 1.94 with a 95% confidence interval ranging from .02 to 3.86. Therefore, hypothesis 3 was rejected. Additionally, in looking at the constructs, the paired -samples t -test for adequate information (see Table 1) revealed that there was a statistically significant decrease in adequate information composite mean scores for the control group of adjunct -only participants fro m the pre -test (M=14.74, SD=3.27) to the post -test (M=13.79, SD=3.05), t (33) =2.10, p= .05 (two -tailed). The mean change in adequate information scores was .94 with a 95% confidence interval ranging from .03 to 1.85. The eta squared statistic (.07) indi cated a moderate effect size. Comparison of means. Though there were few significant results in the t -tests, mean group scores suggested a consistent trend. In all tested cases for both groups, post -test means were lower than pre -test means (see Table 2 ). Lowest means appeared within the construct of connectedness for both the experimental and control groups. This aligns with the findings of previous research with this population (Wicks, 2013). 76 Table 2 Socialization construct group means for pre -test and post -test Construct Group Pre-test Mean Post -test Mean Increase/Decrease Ownership Control Control (Only -Adjunct) Experimental Experimental (Only -Adjunct) 14.91 14.65 15.87 15.63 14.26 13.91 15.61 15.54 Decrease Decrease Decrease Decrease Adequate Information Control Control (Only -Adjunct) Experimental Experimental (Only -Adjunct) 14.85 14.74 15.06 15.08 14.13 13.79 15.03 14.96 Decrease Decrease Decrease Decrease Connectedness Control Control (Only -Adjunct) Experimental Experimental (Only -Adjunct) 12.93 12.38 13.39 13.38 12.83 12.12 13.06 12.96 Decrease Decrease Decrease Decrease Correlations. Additionally, Pearson product -moment correlations were used to explore potential relationships between socialization scores on the pre - and post -tests and on population demographics such as age, gender, years of teaching experience, off -campus status, and employee group affiliations. This aided in determining if group means should be compared holistically or within other certain categorical constraints to inform results. Though several points for consideration were drawn from these tests, as described bel ow, the coefficients of determination were quite small, suggesting a very minor impact from these factors. In terms of the pre -test, correlations did not reveal any significant relationships for gender, age, and years of service. However, a weak, positive correlation was indicated between off -campus status and perceived connectedness (as measured by a composite score on the pre -test socialization scale items related to connectedness), r =.236, n = 77, p < .05. The related coefficient of determination was only 5.6%, so off - 77 campus status may only aid in explaining a very small amount of the variance in participantsÕ scores on the pre -test. This variable did not manifest with a significant score in post -test correlations. In terms of the post -test, correlati ons did not reveal any significant relationships for gender, years of service, or off -campus status. However, a weak, positive correlation was indicated between age and ownership (as measured by a composite score on the post -test socialization scale items related to ownership), r = .231, n = 77, p < .05. So, older participants seemed to have perceived more ownership, while younger participants seemed to have perceived less. Additionally, a weak, positive correlation was indicated between age and adequate information (as measured by a composite score on the post -test socialization scale items related to adequate information), r = .225, n = 77, p < .05. This indicated that older adjuncts perceived themselves as being better informed than did younger adjunc ts. That said, the coefficients of determination for age in relation to both of these scale factors were only 5%, so neither adequate information nor ownership help explain a noteworthy amount of the variance in respondentsÕ scores for the post -test. Add itionally, these variables did not manifest with significant scores on the pre -test. Summary . In summary, all three hypotheses related to research question 1 were rejected. In the case s of the first two hypotheses, t -tests did not illustrate that socialization scores for experimental group participants either increased across time (hypothesis 1) or that their scores were different than the control groupÕs socialization scores on the post -test in a statistically -signficant manner (hypothesis 2) . The third hypothesis was rejected because t -tests showed that the control group socialization scores were not stable and had, in fact, actually decreased in a statistically -significant manner 78 for the construct of ownership across time. Additionally, with a sub -set of control group participants that were only adjuncts (not holding any dual status as hourly staff or administrators), socialization composite scores also showed a statistically -signific ant decrease across time. Research Question 2 The second research question was What themes in dialogue and participation emerged during the experimental group intervention? This question was primarily analyzed utilizing thematic content analysis and LMS u se data. Although the midpoint in the one -month intervention illustrated a record 581 page views in the LMS Moodle Socialwall space and 63 posts, participants only visited the space on an average of 6.63 days in the month. Participants, on average, logge d 50.93 page views and made 6.3 posts or comments in the space across the course of the month -long intervention period. This translates to an average of 12.73 page views and 1.58 posts per week per user. In terms of feature use, participants made fair use of the ability to post (20 posts total) and comment on posts (39 comments recorded ) on the Socialwall. However, they rarely made use of other features common to social media, such as the ability to Òlike Ó one anotherÕs posts or the ability to share links and media. Instead, they gravitated toward activities native to a traditional LMS experience in Moodle, such as posting in forums (which were initiated by a participantÕs request) and participating in the Moodle feedback and glossary activities that were employed by weekly prompts /activities for the gift card drawing s. Socialwall data also revealed that only 23% of the experimental participants (N=7) visited the Moodle Socialwall space with any consistenc y. There was a clear 79 disparity between those with low engagement that had only accessed the space on several days and a subset of 7 users with high engagement that had visited weekly and for a total of 10 or more days in the month. The low engagement user s were split in their tendency to simply pop in on a couple of days and lurk, or to pop in on several days and make a few posts. To further investigate any trends related to highly engaged user activity, the mean composite scores for socialization and its constructs for this subset of users were examined across time through paired -samples t -tests, yielding no significant results. The highly -engaged usersÕ mean scores on the post -test were also compared with the mean scores of the control group through an independent samples t -test, again yielding no significant results. The interesting thing is that, while their engagement was higher by definition of their frequent and consistent visits in the Socialwall space, their page views and posts were actually low er than the participant average collectively. These 7 users had an average of 27.23 page views and 3.2 posts across the course of the month, or 6.81 page views and .8 posts per week as compared to the 12.73 weekly page views and 1.58 weekly posts of the l arger group. In terms of conversational themes, thematic content analysis (Glesne, 2011) was performed with the text of Socialwall interactions to aid in addressin g the second research question (s ee Appendix L). A collective 139 typewritten actions (e.g., posts, comments, edits, etc.) from experimental group participants , were analyzed . 9 of the 24 final experimental group participants, or 37.5%, only participated in the Socialwall in passive ways such as through page visits and views . Specifically, thes e typewritten actions from those participating actively were sourced from all activities initiated in the Socialwall space including 20 Socialw all posts, 39 Socialw all comments , 41 Moodle 80 forum posts, 16 Moodle feedback activity responses , 10 Moodle g lossary activity entries, 7 Moodle g lossary activity comments , and 6 edits to a collective Google D ocument introduced in the Moodle Socialwall space . Participants rarely made use of other features in the space such as the ability to ÒLikeÓ their peersÕ po sts or sharing of hyperlinks and other media, so these kinds of participation were not factored into the analysis. Thematic content analysis was utilized as opposed to other forms of analysis in qualitative research such as conversion analysis, narrative a nalysis, or grounded theory (Glesne, 2011), because of the concrete aims of the question. The goal was to add context to this largely -qualitative report on the potential of the Socialwall tool, not to act as stand -alone data for interpretation. Themes em erging from that data are outlined in Table 3. It is important to note that, due to the richness of comments, comments were coded in multiple ways according to their content. Sub -themes for many of the themes contained comments related to ownership, adequ ate information, and connectedness. 81 Table 3 Themes emerging in Socialwall interactions Theme Sub-Themes Personal Characteristics & Initiative (45 comments, n=16) ¥ Adjuncts need to be active in connecting. ¥ Adjuncts need to pursue necessary information & resources. ¥ Adjuncts need to own oneÕs role in institution & department. Information (75 comments, n=21) ¥ There is a need to foster connection with adjuncts. ¥ There is a need to provide necessary & relevant information to adjuncts. ¥ There is a need to show v alue for adjuncts through inclusion & opportunity. Connection with Colleagues (67 comments, n=19) ¥ There is a need to provide necessary information & feedback to adjuncts. ¥ There is a need to foster connection & support with/of adjuncts. ¥ There is a need to help adjuncts contribute to the departments & institution. ¥ There is a need to value adjuncts through inclusion & opportunity. Technology (55 comments, n=16) ¥ There is a need to provide necessary resources & support to adjuncts. ¥ There is a need to value adj uncts through technology -mediated inclusion & opportunities. Challenges to Onboarding (18 comments, n=11) ¥ There is a need to acknowledge adjunctsÕ diverse experience & commitments. ¥ There is a need to value adjuncts through accessible inclusion & opportunities. Participant quotes were pulled to provide further context for the comments made in each of these 5 thematic areas. These quotes were selected to represent the range in sub -themes that appeared within this larger theme. Pseudonyms are used to protect the identity of the participant. Personal characteristics & initiative. Here, comments called upon the capacities of the individual to share interests and ideas with others. Adjuncts were encouraged by their colleagues in the Socialwall to become a member of, and contributor to, the college community. Specifically, there were three sub -themes for this theme. The 82 first sub -theme suggested that adjuncts need to be active in connecting, as represented by reaching out to others and getting i nvolved in campus events. This is represented by a comment from a user who discusses her own tendency to take initiative in terms of forging peer connections and encourages her peers to do the same by providing examples of what this might look like: ÒOthe r tips for meeting colleagues and feeling connected is to use the adjunct offices, mingle when you are sharing an office with someone, go to meetings like Education on the edge, Gen. Ed. and others, volunteer for the BBQ or the Sweat Shaker, greet instruct or who have the class before you or after you, use the mail/copy room and greet those who are in the mail/copy room with you, go talk to your department head or send them an e-mail.Ó (Jennifer, female, early -30Õs, adjunct for 5 years) The second sub -them e referred to adjuncts Õ need to pursue necessary information and resources by accessing the opportunities and information that exist. This is represented by a comment from a user wh o states that she tends not to take advantage of some of the information a nd resources around her. S he recognizes this as a limitation : ÒWe are always offered resources. Honestly, I do not always take the time to use them. I was asked publicly by the president of the college, [Name Omitted] (Yes, I have been around for thre e administrations) if I got/read the part time adjunct manual. I got it three years after I started at [College Omitted] and had already invested time in figuring my way around, so I had to say no. I did not learn my lesson ÑI still have this same strateg y of finding out what I need, but that could limited me on new possibilities that I have not considered.Ó (Beverly, female, early -50Õs, adjunct for 14 years) The final sub -theme indicated that adjuncts need to own their role in the institution and depart ment by seeking connections with others in their discipline and helping to drive departmental directions and college initiatives. This is represented by a comment from a user who discusses that there is opportunity to become involved in her department : ÒIÕm lucky, I guess, being in the [omitted] DepartmentÉWe meet regularly and the opportunity is there to get involved. I have found that I often have to take the initiative to become a part of the process. It works 83 for me, but it may not work for everyone .Ó (Corinne, female, late -50Õs, adjunct for 7 years) She mentions taking personal ÒinitiativeÓ as important to becoming involved in departmental processes. Information. A second overarching theme in participants Õ comments on the Socialwall concerned access to, and organization of , information along with its perceived relevance to their work . There were also three sub -themes related to this theme. The first sub -theme related to a need to foster connection with adjuncts when we think about informatio n in terms of ensuring that they are informed . This is represented by a comment from a user who discusses how adjuncts are sometimes left unaware of pertinent information related to the college . This lack of information sparks conjecture : ÒI offer this a necdote not just for itÕs distracting appeal : -) but also to illustrate that in the absence of information, most will do their best to fill in the missing information. An organization then, should be cognizant of what it is letting ÔnewÕ employees fill in ...Ó (Gerry, male, mid -30Õs, adjunct for 15 years) The second sub -theme in participantsÕ comments indicated that they believed the college should provide relevant and up -to-date information to adjuncts that directly apply to their roles and their unique needs . This is represented by a comment from a user who discusses how events like orientation for adjuncts may be helpful at first but, if they are not revised to meet participant needs over time, they become meaningless: ÒAs a new adjunct, the first two or three years that I worked for [College omitted], I attended orientation and considered it an exercise in protocol and setting an example, but after awhile I realized that it was repetitious and found myself skipping the breakout sessions that I had already taken.Ó (Corinne, female, late -50Õs, adjunct for 7 years) 84 The third sub -theme in participantsÕ comments related to information organization and access was that they believed the college should demonstrate that it value s its adjuncts through inclusion and opportunit ies that acknowledge their schedule constraints . This is represented by the following comment from Renny, who expresses his belief that adjuncts with other daytime commitments are often excluded from college even ts. He suggests leveraging technology to provide more opportunities for inclusion: ÒAdjuncts are not on campus 8-5 Mon -Fri and some have other places of employment. [ College o mitted] could remove an obstacle of physical presence to attending these meetings by using old technology such as teleconferencing so adjuncts can participate in meetings regardless of their physical location. Ó (Renny, male, late -40Õs, adjunct for 11 years) Connection with colleagues. Another overarching theme in particpipantsÕ comments on the Socialwall concerned their feelings of connection, or lack thereof, the potential value of such connections, and methods for cultivating connections. There were four sub -themes related to this theme. First, comments emphasized the importance of connecting with oneÕs supervisor or ful l-time. This is represented by a comment from a user who expresse s being so disconnected and without feedback as an adjunct that he had no knowledge of his supervisor, which left him feeling uncertain of whom to go to for information and feedback: ÒI didnÕt even know who my ÔsupervisorÕ was for two years, let alone wher e to go for advice.Ó (Steven, male, early -60Õs, adjunct for 6 years) A second sub -theme expressed in the comments was the lack of connection or isolation on the job for some adjuncts . This is represented by a comment from Anya relating that her daily exp erience at the college was limited to teaching and student interactions without other interactions that would promote connection or Òfeeling valued :Ó 85 ÒI come in for office hours, teach the class, and leave. Therefore, there are no comments to give regardi ng support or feeling valued/heard.Ó (Anya, female, late -40Õs, adjunct for 3 years) On the other hand , some adjuncts felt they were connected to colleagues in their department and the institution . This is represented by a comment from one participant, Sam, who indicated how his department had taken active steps to welcome and include him, readily soliciting and hearing his opinions: ÒOne thing I can say, however, is that I enjoy teaching here very much, and I appreciate my colleaguesÕ welcoming me into their circle. IÕve never felt more welcome, or that my opinion mattered, as much as I do here.Ó (Sam, male, mid -30Õs, adjunct for less than 1 year) Fourth, comments suggested methods for cultivating connections, such as fostering relationships at departmental or college events a nd seeking input on college processes. This is repr esented by a comment from Beverly detailing how she feels unvalued and not included by her full -time faculty peers e ven at college -sponsored events: ÒSince we have plenty of chances for connecting in our department, on in -service days I can sit with and get to know others. Those who do not meet with their departments could make it a point to do so during lunch etc. It seems that some of the full -timers could care less abo ut this though (I see them sitting alone, working on their classes, grading etc.).Ó (Beverly, female, early -50Õs, adjunct for 14 years) Technology. Technology -related comments included questions and requests for support, general statements of interest, and discussion about how technology might aid connection and learning. There were just two sub -themes related to this theme. The first sub -theme recognized a need to provide necessary resources and support to adjuncts when it comes to technology, which c ould include things ranging from proper training and equipment to technical support representatives . This is represented by a comment 86 from a user discussing her difficulties in implementing iPads with her students without at-hand technical support: ÒI cou ld have used some help from a tech person to decrease student frustration.Ó (Beverly, female, early -50Õs, adjunct for 14 years) The second sub -theme indicated a need to value adjuncts through technology -mediated inclusion and opportunities such as events that are also offered remotely or asynchronously online . This is represented by a comment from a user discussing how numerous participants cannot attend campus -wide events like orientation and professional development days, but how there may be relatively simple ways to include more users through synchronous and asynchronous technologies: ÒAlso, for those who can not make it physically to Orientation and PD, why not use Skype or some other online equivalent? If needed we could do a registration or make vi deos to upload for future reference.Ó (Jennifer, female, early -30Õs, adjunct for 5 years) Challenges to onboarding. Here, comments focused on the challenges to onboarding adjuncts due to the various roles they play, the fact that many work at various institutions or other venues, and the experience they bring with them. There were also two sub -themes related to this theme. A first theme in the comments was that adjunctsÕ diverse experience and commitments may impact their ability to b e involved in college life and events, even if they are willing . This is represented by a comment from a user detailing how a reduction in the credit hour assignments available to her at the college created a need to supplement income with other teaching commitments, which then decreased her access to campus events and opportunities: ÒAs the courses reduced in size, it created a gap that needed to be filled with adjunct teaching positions at other colleges/universities, which in turn minimized the amount o f time that I could be involved in [College 87 omitted] Õs extracurricular and on campus activities.Ó (Corinne, female, late-50Õs, adjunct for 7 years) The second theme was that inclusion and opportunities to be involved should be more accessible , which might include advance notice of activities, flexible options, and technology -facilitated opportunities. This is represented by a comment from a user discussing how his combined life commitments left him with little time to engage in socialization opportun ities, so he needed advance notice to carve out time: ÒMy plate is rather full for the next few weeks. I would like to know the date/time that is decided so I could attend if it does not conflict with my work or volunteer work.Ó (Jim, male, early -60Õs, ad junct for 5 years) Summary. In regards to the second research question, there were multiple themes emerging related to participation during the intervention. Some of these themes were related to use, with statistical data from the LMS illustrating that some users engaged with the Socialwall space with weekly regularity (n=7) throughout the intervention, while some visited it much less frequently and sporadically (n=17). Additionally, those who engaged with more regularity had a notably lower number of page views and posts tha n the group collectively. O nly 15 users (62.5%) made posts or other active attempts to outwardly engage with peers in the Socialwall space. Other themes were related to the conversational content generated by participants in the space. Major themes in the So cialwall comments from this small subset of verbal experimental group participants (n=15) related to: ¥ personal characteristics and initiative, by way of adjuncts taking ownership of opportunities to connect, learn, and contribute; ¥ information, in terms of providing relevant and accessible information that meets the unique needs of adjuncts; 88 ¥ connection with colleagues, through direct links with colleagues and opportunities to contribute and receive support; ¥ technology, including technology tools , training, a nd support along with leveraging affordances of technology to increase access ; ¥ and, finally, challenges to onboarding, including recognition of the diverse experiences and life commitments of adjuncts and providing flexible opportunities for them to learn and engage. There were clear ties in a number of these themes and their related sub -themes that aligned with the constructs of socialization ( ownership , adequate information , and connectedness ). In addition, themes related to technology and onboarding cha llenges emerged. These issues inform this model of socialization and how it was operationalized, as they were not originally indicated as issues of interest within this framework. Research Question 3 The third research question was , ÒRQ3: What benefits, if any, did the experimental group perceive that the intervention provided to support their socialization? Ó This question was explored in several ways. First, it was explored through participant responses to Likert -type questions regarding the interventi onÕs impact on socialization. It was also explored thematically through both open -ended survey questions about the intervention experience along with Socialwall intervention comments regarding the experience. Likert -type s cores. Intervention participan ts (n=30) were asked three Likert -type questions on the post -test survey regarding their experience in the Moodle Socialwall pilot. Mean scores related to the perceived impact of participation on ownership were 89 notably lower than mean scores related to th e perceived impact of participation on access to information or connectedness . The majority of the experimental group agreed that the intervention supported their socialization, particularly in terms of the constructs of exposure to adequate information and connectedness, though they were neutral or uncertain that the intervention supported the construct of ownership. Response means and standard deviations are shared in Table 4. Table 4 Participant perceptions of the SocialwallÕs support for s ocialization Question Mean Score (1=SD to 5=SA) % of Responses (N=30) Participation enhanced my sense of ownership for my primary academic department. 2.87 (SD=.81) Strongly Disagree or Disagree: 29% Neither Agree Nor Disagree: 55% Agree or Strongly Agree: 16% Participation exposed me to information that is relevant to my service as an adjunct. 3.32 (SD=.83) Strongly Disagree or Disagree: 16% Neither Agree Nor Disagree: 32% Agree or Strongly Agree: 52% Participation supported my sense of connectedness with my colleagues. 3.42 (SD=.92) Strongly Disagree or Disagree: 16% Neither Agree Nor Disagree: 29% Agree or Strongly Agree: 55% Open-ended survey questions. Experimental participants were asked several open -ended survey questions regarding whether or not they felt they benefitted from their exposure to the intervention. Responses were explored through thematic content analysis. In general, questions did not effectively elicit specificity in participant responses and many of the t hemes in participant response carried across all four questions. For instance, many responses were brief, consisting of one-word answers or phrases such as ÒNone,Ó ÒSee above,Ó ÒGood info,Ó or ÒGot to know some other faculty.Ó Below, I discuss response trends related to each o f these four questions 90 separately. Next, I discuss what I see as the overarching themes represented in these res ponses. The first question asked, What do you feel you gained from the Socialwall experience, if anything? Of the 24 experimental group participants that took part in the Socialwall and completed the post -test, 22 responded to this question. Some of the responses contained multiple themes, so they may be counted more than once in the following themes. First, 5 out of the 22 participants stated they received no benefits from the experience. 10 participants responded that they gained a sense of connection with peers. 6 participants said they gained increased knowledge and expe rience with social media tools. 5 named helpful information and perspectives as a benefit. Lastly, one participant talked about seeing a sense of recognition for opinions occur in the space. The second question asked, What features of the Socialwall experience did you feel were most important in supporting your sense of ownership for your primary department, if any? Please describe. 18 participants responded to this question. An overwhelming subset of participants (11) indicated that there were no perceived features of the Socialwall that supported ownership. Another 4 indicated that they had not participated enough to receive any benefit s related to their sense of ownership for their primary department . Of those few participants that did feel the Socialwall features supported their sense of ownership, 3 mentioned information and diverse perspectives. Another 2 discussed connection, including bridging social capital. Finally, one respondent was pleased that she Òwas not the only one out there that didnÕt know anybody.Ó 91 The third question asked, What features of the Socialwall experience did you feel were most important in supporting your access to information relevant to you r service as an adjunct faculty member, if any? Please describe. 17 responses were received here. 6 participants reported no benefits related to adequate information. However, the other responses covered a variety of areas. 4 participants noted the val ue of information shared in the space while another 3 noted its benefits for communicating with peers. 2 participants noted that it had potential to support access to information if better organized and 1 appreciated that ease of access of the space. Ano ther participant mentioned that it provided information on a new tool and one more discussed that the experience helped reinforce existing viewpoints sheÕd held. Lastly, the fourth question asked, What features of the Socialwall experience did you feel we re most important in supporting your sense of connectedness to your colleagues at [College omitted] , if any? Please describe. 18 responses were received. 7 participants noted that collaboration and sharing of perspectives with colleagues supported their sense of connectedness. 5 listed connection such as putting names and faces together as a benefit. 4 mentioned that conversations specific to teaching were particularly engaging for them. 3 observed no support for connectedness in the experience. 2 sp oke of the ability to crowdsource in the space for questions and answers. Finally, one participant mentioned feeling affirmed and valued when her name was used by peers. The 75 open -ended survey responses were then combined in one document for thematic co ntent analysis (see Appendix L). Responses were considered collectively due to the themes that seemed to carry through responses regardless of the question or 92 socialization construct pinpointed in that question. With the combined approach to analysis, fi ve overarching themes were identified (see Table 5). Table 5 Themes emerging in post -test regarding Socialwall for s ocialization Theme Examples Connection ¥ Connection/ collaboration with peers. ¥ Receipt of, or providing, s upport to/ from peers. ¥ Exposure to diverse perspectives. None ¥ There were no noted gains. ¥ There were no important features. Information ¥ Relevant information was shared. ¥ Information was easily accessed. Lack of Participation ¥ Acknowledgement of oneÕs own failure to engage. Technology ¥ Expos ure to, and practice with, a new tool. ¥ Greater understanding of social media tools. These themes appeared to echo the Likert -type scores related to the constructs of socialization. Participants acknowledged benefits from the experience related both to connection and to information, though they struggled to name benefits related to the construct of ownership. Additionally, par ticipant quotes were pulled to provide further context for each of the five thematic areas of response. Several q uotes were selected to represent these categories . However, these quotes are brief because many of the responses were merely several word phra ses, as described above. Pseudonyms are again used to protect the identity of the participant. Connection. Comments here related to connection, sharing, and support with and from peers. Examples of this theme included connection and sharing with peers on both personal and professional levels , support from peers in terms of camaraderie and emotional support , and exposure to diverse perspectives that broadened oneÕs perspective and sparked further consideration . In this comment, a participant indicates how personal recognition from peers helped her feel valued: ÒWhen my name was used in a post by a 93 colleague I felt particularly affirmed, connected and valuedÓ (Veronica, female, early -50Õs adjunct for 15 years) . Here, a participant indicates a value in exposure to diverse perspectives from others: ÒReviewing comments from my colleagues and considering their view points if different from my own Ó (Darryl, male, late -40Õs, adjunct for 2 years ). None. There was a c lear subset of users with comments suggesting that they receive d no value from the experience. They indicated either that there were no noted gains, in terms of the experience they had in the Socialwall, or that there were no important features , in terms o f the Socialwall software plug -in. In this comment, a participant indicates a lack of benefit and disengagement caused by difficulty navigating the interactions: ÒNone. It was bewildering to try and follow ing the myriad of conversationsÓ (Aaron, male, early -30Õs, adjunct for 6 years) . Here, a participant simply communicates that he did not perceive any helpful features in the software: ÒNo features stand out Ó (Jim, male, early -60Õs, adjunct for 5 years) .Ó Information. These comments related to the acc essibility and relevance of information shared. This included comments on the accessibility of the online space and that relevant , or beneficial information shared within it . In this comment, a participant appreciates the ease of locating the Socialwall space as a solution native to the institutional LMS : ÒIt was easy to find Ó (Sabrina, female, early -30Õs, adjunct for less than one year) . Here, a participant talks about the value and application she saw in a specific conversation and how it reinforced he r own beliefs: ÒI found the discussion related to participation at professional development/orientation days relevant to understanding views of others, wh ich reinforced my views of same Ó (Veronica, female, early -50Õs adjunct for 15 years) . 94 Lack of participation. Participant comments related to this theme recognized a failure to engage with the Socialwall tool and experience at a level that might reap benefit. In this comment, a participant offers an acknowledgement of her own lack of engagement in the experience , which impacted the benefits she received: ÒNone. Perhaps if I had participated mor e, my answer might be different Ó (Katrina, female, late -40Õs, adjunct for 5 years) . Here, a participant notes a lack of engagement due to technical difficu lty with the software: ÒIÕm afraid I did not do much with it. Frankly, it was a b it difficult for me to navigate Ó (D ebra, female, early -60Õs, adjunct for 14 years ). Technology. Lastly, p articipants commented on exposure to, and practice with this new tool as well as a greater understanding of social media. In this comment, a participant discusses her newfound knowledge of the Socialwall technology and how it may impact her future practice: ÒLearned about another social media tool that may be beneficia l in teaching and communicating with students and peers Ó (Katrina, female, late-40Õs, adjunct for 5 years) . In this comment, a participant discussed her increased awareness of social media and how it functions: ÒI felt that I gained a better understandin g of using social media for the classroom Ó (Rosa, female, early -60Õs, adjunct for 3 years) . Socialwall comments. Another level of analysis completed in regards to RQ3 was the extraction of the 26 Moodle Socialwall typewritten actions (e.g., comments, posts, etc.) out of the total 139 actions that related to the Moodle Socialwall experience or tool. These comments were not in response to any specific question posed by the researcher, just posts on this topic that ap peared on the Socialwall or in the forums and other activities used within the Socialwall space. It is somewhat difficult to compare and 95 contrast this data to the quantitative and qualitative survey data previously presented because participants were not specifically asked to interact about the Sociawall experience within the Socialwall space. Though a Google Document was initiated to record their comments on how the software may be improved, they were not directed to talk about if or how the tool had aid ed their socialization in relation to its three constructs. That said, a n abbreviated version of thematic content analysis was performed to identify three emerging themes discussed below. Comments are again included to provide context for these responses along with pseudonyms to protect the participantsÕ identities. First was the theme of purpose (12 comments , n=8). In these comments, participants either demonstrated uncertainty about the purpose, utility, or value of the Socialwall space , or they made suggestions for how it could be used . In this way, it was clear that participants felt no unified sense of purpose for the space or for how this social media tool should be used. This may help explain the survey responses indicating that participant s had lack of participation or those suggesting no perceived benefits from the experience. It may also help explain why there was a perceived neutral support for ownership on Likert -type scores and why participants failed to fully engage with the Socialwa ll. Here, one participant comment ed on his concerns related to the utility and value of the Socialwall, which he believes might be mediated through careful design: ÒFor me, the primary issues relate to usefulness, connectivity, and site organization/featu res. If these things are done well, it might draw faculty in and continue to grow. If not, it becomes just another social media fluff site that is in competition with 100 other sites where people say a lot about very little and useful information is diff icult to find.Ó (Stan, male, late -50Õs, adjunct for 33 years) 96 Another participant offered insight into how the Soicalwall might be used within departments for curricular information -sharing: ÒI like the idea of keeping the space open for departments so we instructors can share information back and forth that will help us better track how well the students are moving from class to class. Instructors could also share learning materials pertinent to the departments.Ó (Anya, female, late -40Õs, adjunct for 3 years) Second was a theme related to positivity and potential (13 comments , n=12). Many comments made in the Moodle Socialwall intervention space about the experience were positive ( 13 comments ). Additionally, comments expressed the potential value (12 comments ) of the Socialwall space across time. This theme may indicate support for Likert -type scores and survey responses in which participants perceived that the Socialwall did support their socialization by providing connectedness and access to info rmation. In general, comments echoed the benefits reported on the survey in terms of how the Socialwall or social media did support or might support socialization (e.g., connection and sharing with peers, support from peers, exposure to diverse viewpoints , access to relevant information). Here, one participant expresses positive thoughts related to her experience and engagement in the Socialwall spaceÓ ÒI enjoyed the Socialwall and visited the site almost daily to see what was posted that was new.Ó (Carr ie, female, mid -60Õs, adjunct for 6 years) Another participant offered speculation on how the Socialwall might support socialization in the future, particularly for online instructors with little connection to the physical campuses and the opportunities a nd peers serving there: ÒI think the Moodle Socialwall is a great new adventure toward encouraging adjunct engagement and feeling more involved, especially those of us that only teach online courses.Ó (Corinne, female, late -50Õs, adjunct for 7 years) 97 Third was the theme of functionality (13 comments , n=9). In these comments, participants discussed the features of the Socialwall software, how they functioned, and how effective they were for their intended purpose. These software -related comments could be related to the perceived lack of benefit expressed by some in survey responses who indicated difficulty following conversati ons in the space and a lack of engagement. Here, a participant discusses challenges asso ciated with navigating conversations in the Socialwall: ÒI agree that it is somewhat difficult to keep track of conversations and responses taking place, as it seems too busy.Ó (Sam, male, mid -30Õs, adjunct for less than 1 year) Another participant expre ssed a technical request that she felt would enhance the experience of participating in the Socialwall space: ÒI think the posting box needs to be a little bigger.Ó (Rosa, female, early -60Õs, adjunct for 3 years) Summary. In relation t o the third resea rch question, a small subset of verbal participants perceived that the Socialwall intervention support ed their socialization in a variety of ways. In terms of trends, Likert -type data revealed over 50% of experimental group participants either agreed or strongly agreed that the intervention supported their needs for adequate information and connectedness, though responses related to ownership illustrated that 71% of experimental group participants had neutral or conflicting feelings abou t whether or not the intervention offered support for this construct. Post -test survey responses indicated five themes in how participants thought the intervention supported their socialization or failed to do so (connection, none, information, lack of pa rticipation, and technology) along with related sub -themes. These post -test themes echoed the SocialwallÕs cap acity to support connectedness and 98 information sharing that were indicated in the Likert -type survey data. They also echoed the quantitative neu trality about its ability to support ownership because participants struggled to report benefits related to that construct, with some noting their own lack of ownership in participating or noting no benefits from the experience. Lastly, in Socialwall typewritten actions that were specifically about the Socialwall, three different themes related to participantsÕ perceptions of the tool and the experience bubbled up (purpose, positivity /potential, and functionality). The notes on poten tial and positivit y expressed by participants about the experience offered support for the survey data (both qualitative and quantitative) suggesting that they perceived support for socialization in the space in terms of connectedness and access to information. The questio ns about purpose, functionality, and the notes about uncertainty echoed some of the concerns noted in the survey comments and also in the neutrality on the survey data related to ownership . In looking across survey data and Socialwall comments, participan ts reported that the Socialwall can aid socialization through its capacities for supporting connection and sharing amongst peers, for peer support, for exposure to diverse viewpoints, and for easy access to relevant information. For instance, as will be discussed further in chapter 5, the comments captured from typewritten actions participants made in the space indicated how they drove the conversation with personal interests, how the conversation included both targeted communications and feedback and how it promoted peer modeling, and how participants shared opposing perspectives and offering one another support. 99 Conclusion To summarize the analysis, both quantitative and qualitative data were gathered and inspected in relation to three hypotheses and tw o research questions. Though this study utilized the convergent parallel mixed -methods design, the primary goal of this research was a quantitative exploration of the potential impact of the new Moodle Socialwall format plug -in to impact perceptions of so cialization for its adjunct faculty users. The qualitative data served as a secondary means for exploration, providing further context surrounding quantitative results. A small verbal subset of experimental group participants perceived that the Socialwall intervention support their socialization in a variety of ways, though their actual use of the space was fairly limited in terms of those who made regular visits to the space (n=7) and those who entered any comments or posts in the space (n=15) . In terms of the first research question and its hypotheses, all three hypotheses were rejected. The first two hypotheses proposed that intervention participants would report higher average perceptions of socialization than their control group peers at the end of the study and also that intervention group participantsÕ perceptions of socialization would be higher at the end of the study. The t -test results failed to support these hypotheses. The third hypothesis proposed that control group participants wo uld report consistent perceptions of socialization across time. However, ownership scores decreased in a statistically significant way for the control group across time. Additionally, once the subset of participants holding dual status as adjuncts and as administrators or support staff were removed, the t -test revealed a statistically significant decrease in socialization means from pre -test to post -test. This resulted in the rejection of 100 that last hypothesis, th ough the effect size was small. Also, wit h the removal of those with dual roles in other employee groups, there was a statistically significant increase in adequate information composite scores for the control group from the pre -test to the post -test, with a moderate effect size. Additional quant itative analysis included comparison of means for socialization and its three constructs and Pearson product -moment correlations to examine potential relationships among scores and other data. Means decreased on the post -test in all cases for both groups. Lowest means appeared within the construct of connectedness for both the experimental and control groups, which aligns with the findings of previous research with this population (Wicks, 2013). In terms of the pre -test, correlations did not reveal any si gnificant relationships for gender, age, and years of service. Pre -test correlations indicated a weak, positive correlation between off -campus status and perceived connectedness. On the post -test, correlations indicated a weak, positive correlation betwee n age and ownership and also between age and adequate information. So older adjuncts perceived more ownership and they theyÕd received more adequate information than did younger adjuncts. However, none of these correlations appeared across both pre-test and post -test. In terms of the second and third research questions, results were drawn largely from thematic content analysis. The second research question asked about the themes in intervention participation. I listed and described the 5 primary themes that surfaced during Moodle Socialwall interactions (personal characteristics & initiative, information, connection with colleagues, technology, and challenges to onboarding) along with their sub -themes. Most of the themes and their sub -themes suggested n eeds aligned with the 101 three constructs of socialization in the model used in this study (ownership, adequate information, and connectedness). I also explored LMS use data, finding that participants viewed the space an average 6.63 days across the one -mont h intervention, logging an average 12.73 page views and 1.58 posts per week. Though many users had low engagement, 23% of participants (a subset of 7 users) were highly engaged in terms of their consistency in accessing the space. These 7 users visited w eekly and on 10 or more days during the month. However, they had only 6.81 page views and .8 posts per week on average, which was lower than the group collectively. Means for this subset of users was explored, finding no significant results through t -tests. Lastly, the third question asked about participantsÕ perceptions of the interventionÕs impact on their socialization. Both quantitative data along with open -ended survey post -test responses and Socialwall conversation data were thematically analyzed t oward this aim. Likert -type data means suggested that ownership was the area in which participants saw least impact , though they did perceive that the Socialwall supported connectedness and adequate information . Open -ended survey responses from the post -test indicated five themes (connection, none, information, lack of participation, and technology) with conne ction being the most frequently noted benefit. Many of these themes and their sub -themes also aligned with the three constructs of socialization (ownership, adequate information, and connectedness). Though participants identified benefits related to connectedness and access to information, they had difficulty naming benefits related to ownership, which echoed the Likert -type data. Additionally, com ments and posts about the Socialwall made in the Socialwall space were analyzed thematically, resulting in three different themes (purpose, positivity/potential, and 102 functionality) that described how users viewed the space immediately and across time. Tho ugh none, as in no noted gains (n=15), was a theme of user perceptions on the combined post -test questions the Socialwall data was quite positive, illustrating a large degree of positivity and potential (13 comments from 12 participants ), though uncertainty about the tool Õs purpose did exist (13 comments from 9 participants ). In looking across survey data and Socialwall comments, a small verbal subset of experimental group participants reported that the Socialwall can aid socialization through its capacities for supporting connection and sharing amongst peers, for peer support, for exposure to diverse viewpoints, and for easy access to relevant information. However, these responses must be considered in the contex t of use data. For instance, if only 7 users visited the Socialwall space with any regularity, how well can their experiences really translate to that of the vastly l arger population? Also, there were experimental group participants that did not particip ate with any regularity and may not have even made posts in the Socialwall space, but still offered post -test commentary on the experience. Can they comment with relevance for the larger population on an experience in which they remained passive participa nts ? In the next chapter, I will further discuss and interpret study results. Qualitative data will be referenced for context in synthesizing quantitative results. Emerging patterns will then guide the recommendations for future research. 103 CHAPTER V Discussion and Implications The goal of this research was to explore the potential of social media, specifically the new Moodle Socialwall format plug -in, for supporting adjunct faculty socialization. To achieve this, a design resembling the nonequivalent (pre-test and post -test) control -group design was employed. Below, I discuss the three related research questions and their results along with potential explanations for those results in the context of my research methods, my observations as the research er, and the previous empirical literature in these areas. Then, I outline some of the limitations in design, sampling, and procedures for this study, which lead to recommendations for future research and practice. I conclude with major takeaways, focusin g on the need for further work related to institutional use of social media for socialization in higher education. Overall, I found that adjunct faculty participating in a social media intervention facilitated via the Moodle Socialwall did not report high er perceptions of socialization than their peers after the month -long intervention. I also found that only a minority of adjuncts inv ited to participate in the intervention actually participat ed in a consistent manner . Specifically, just 7 out of 24 fina l experimental group participants visited the space on a weekly basis . Additionally, just 15 participants made posts in the Socialwall space, failing to make use of other features of the Socialwall that are common to social media spaces (e.g., the ÒLikeÓ feature, the ability to share hyperlinks, and the ability to message individually with peers). Adjunct participation on the Socialwall demonstrated conversational themes related to the three constructs of socialization ( ownership , adequate information , and 104 connectedness ). For instance, their conversations covered themes such as the need for adjuncts to take personal initiative ( ownership ), the need to provide adjuncts with relevant information ( adequate information ), and a need to foster connection wi th colleagues ( connectedness ). However, due to the fact that just 15 participants engaged in posting and commenting in the space, these results cannot be generalized to the larger population. Related to this, a djunct participation demonstrated the need t o consider the applicability of this theoretical frame for this phenomenon and the design of such an intervention and the implementation of such a tool to foster increased participation. For instance, some studies suggest that it may take closer to two mo nths to change behavior (Lally, van Jaar sveld, Potts, & Wardle, 2009). Additional themes emerged related to the challenges of onboarding and calls for technology support and use of technology to provide adjuncts with more opportunities for inclusion and involvement in the institution. Despite the lack of increases on self -reported socialization scores , the majority of the very small, verbal subset of those participating in the intervention agreed or strongly agreed that the participation in the Socialwall supported their connectedness to colleagues and access to adequate in formation to perform in oneÕs job role. Aspects of the Socialwall that they perceived as supporting access to connectedness were the ability to connect and share amongst peers along with the ability to provide support to peers through posts, comments, and other elements in the space. Aspects of the Socialwall they perceived as supporting their access to information were the ability to easily access information that was relevant to their role along with gaining exposure to diverse viewpoints though sharing of opinions, approaches, experiences, and content and links with peers . Overall, intervention participants were more neutral when considering if the 105 Socialwall experience supported their sense of ownership. In general, they had difficulty naming element s of the experience or software features that supported this construct. This may be due to the fact that little direction was provided in terms of how they should interact in the space because I wanted to see what types and themes of interaction occurred organically in a social media environment. Participants noted a lack of clarity on the purpose for the tool in their comments in the Socialwall space. Below, I further discuss findings related to each of the three research questions, exploring how these findings intersect with the literature review and conceptual framework presented in Chapter II. Then, I detail potential limitations of this research and its findings related to the design and methods. Lastly, I close with recommendations for future research and practice in these areas. Research Question 1 The first research question ( Do adjunct faculty participating in the Moodle Socialwall intervention report higher perceptions of socialization than t heir peers? ) was analyzed quantitatively with three related hypotheses. The means and hypotheses of this study offered unexpected results in that all hypotheses were rejected and all means (socialization and it s three constructs: ownership, adequate infor mation, and connectedness) decreased for all groups. Below, I outline these hypotheses along with the results of related analyses, and offer discussion on why I believe these results were attained. Hypothesis 1. Adjunct faculty participating in the Moodle Socialwall intervention will report higher average perceptions of socialization than their peers. This hypothesis was rejected because independent samples t -tests did not illustrate a 106 significant difference in m ean post -test scores across the experimental and control groups. All participants on average had less positive perceptions of their socialization both overall and in terms of its constructs (ownership, adequate information, and connectedness) across time. Means for connectedness were the lowest means for both groups on both pre -test and post -test. This aligns with the perceived need for further connectedness expressed by this population in previous research (Finucane & Algren, 2007; Shannon, 2007; Wicks, 2013). This suggests perhaps that adjuncts simply struggle with a sense of connectedness across the board in their positions due to the non -traditional appointment model and other life obligations and demands. Though some might interpret these decreased means scores as a reason to discount the value of the Moodle Socialwall in supporting socialization and its constructs in this model, the results are undoubtedly impacted by elements of design. Though it may be the case that the intervention did not have an impact on socialization, it may also be the case that the sample was too small to accurately gauge impact. Likewise, it may be that the intervention time period or that the scaffolding for tool use were too brief for experimental group participa nts to incorporate the tool in meaningful ways in their practice and to experience its benefits for socialization. Experimental and control group means were very close at the outset of the research , so the lack of a significant difference at the end is not tota lly surprising given the limitations of these methods, which are discussed in more detail later in this chapter . Hypothesis 2. Adjunct faculty participating in the Moodle Socialwall intervention will report higher average perceptions of socialization foll owing the intervention. This hypothesis was also rejected because paired samples t -tests did not 107 illustrate a significant difference in mean scores from pre -test to post -test. Again, I would have to mention the potential explanations noted for hypothesis 1. The small sample size may have impacted the ability to gauge impact. Likewise, the intervention time period and scaffolding for tool use may have been too brief for experimental group participants to regularly use the tool and reap its benefits. As n oted in the methods, the participants were asked to make use of the tool in ways that seemed most meaningful to them with little further direction, other than the sharing of conversational and activity prompts two times per week. This procedure was used t o explore organic benefits that might develop from exposure to the tool as opposed to exploring the impact of specific intervention designs that administration might employ in this space. Also, as noted in the analysis, only around a quarter of experiment al participants (n=7) engaged with the Socialwall on a weekly, regular basis. When they did engage, they only made use of the post and comment features on the Socialwall along with posts in associated forums. They did not make use of other features commo n to social media experiences such as the ability to ÒlikeÓ the posts and comments of peers, the ability to share hyperlinks, and the ability to message directly to peers. About two thirds of participants (n=15) made use of the comment and post features. So, it seems there was not enough activity from enough participants over enough time to impact them in a way that would be visible through the survey. Hypothesis 3. Adjunct faculty not participating in the Moodle Socialwall intervention will report consistent perceptions of socialization across time. This hypothesis was rejected for several reasons. Paired samples t -tests illustrated that there was a statistically significant decrease in ownership scores from pre -test to p ost -test for 108 the control group. The notable decrease in ownership scores could be due to group assignment. For instance, several participants who were assigned to the control group had originally expressed interest in being part of the experimental group so that they could help trial the Moodle Socialwall. However, due to the randomized assignment procedures, that was not possible. The withholding of the treatment could have impacted their perceptions of ownership across time. Creswell (2014) discusses how internal validity can be compromised by compensatory rivalry, when the control group feels devalued because they do not receive access to the treatment. In the model for socialization used here, the construct of ownership is represented by four Liker t-type questions, one of which refers to how valued the participant feels in the work environment. So, if the participants were impacted by compensatory rivalry, it could also have impacted the mean scores for the construct of ownership. Also, it is quit e likely that adjuncts collectively struggle with a sense of ownership , in terms of feeling valued in their work environment, due to the variable nature of their appointme nts. For instance, research has indicated conflicting data regarding adjunct faculty satisfaction with teaching load, salary, and job stability (Kramer, Gloeckner, & Jacoby, 2014). Though the control group mean scores for this construct decreased significantly, the experimental groupÕs mean scores also decreased. When refining the group means to only those holding adjunct status at the institution, the control group also reported significantly lower scores at the end of the study for overall socialization means along with significantly lower means for the construct of adequate information . One reason for the decreased scores may have been that the question prompts on the pre - and post -tests allowed participants to think back on 109 their socialization experiences and any related challenges across the course of the one -month experimental timef rame, which was ultimate ly captured by the post -test. Pearson product -moment correlations. Additional calculations suggested a weak, positive correlation on the pre -test between connectedness and those that only taught off -campus. Though this did not a lso manifest on the post -test, it is understandable that participants that are removed from the physical campus may feel less connected. It could be that the intervention mediated this a bit for some of the off -campus participants so that it did not appea r on the post -test. It could also be that the opportunity to take part in this study and the compensation provided for doing so aided some off -campus participants in feeling more connected to campus activities. On the post -test, there were weak, positive correlations related to age. The correlation between age and ownership suggests that older participants felt more ownership of their roles while younger participants felt less ownership. Additionally, the correlation between age and adequate information suggests that older participants feel better informed than do younger participants. This makes sense because it may be true that younger individuals are career building and constructing their professional identities while older individuals may have alrea dy attained these developmental milestones and may have served at the institution for several years. Career seekers may also have higher expectations for ownership and information than do those working toward retirement or holding supplemental careers. T hough these correlations were not apparent in the pre -test, it could be that the reflection involved in the surveys aided participants in expressing these thoughts on the post -test. 110 Research Question 2 The second research question ( What themes in participation emerged during the experimental group intervention? ) was analyzed using data from the Socialwall intervention and analyzed through thematic content analysis. Below, I detail some of the noteworthy statistics and user characteristics that may account for these statistics. I also list emerging themes from comments and posts in the Socialwall space and discuss how these relate both to the constructs of socialization and to the affordances of social media. Socialwall usage . Participants viewed the space an average of 6.63 days across the one -month intervention, logging an average of 12.73 page views and 1.58 posts per week. This was lower than expected , and that is likely due to some of the factors previously mentioned, such as the brief timefr ame users had to incorporate the tool into their routines and the limited direction users were given on how and why to interact in the space. However, some research suggests that this is a fairly average participation level related to online community con tributions (Schellen s & Valcke, 2005). Though many users had low engagement, 23% of participants (a subset of 7 users) were highly engaged in terms of their consistency in accessing the space. These 7 users visited weekly and on 10 or more days during th e month. However, they had only 6.81 page views and .8 posts per week on average, which was lower than the group collectively. Means for this subset of users was explored, finding no significant results through t -tests. Demographic factors such as age, gender, years of service, teaching discipline, and previous social media use were also considered related to these 7 highly engaged users. However, no demographic commonalities or trends were discovered that would help explain their engagement with the So cialwall when compared to their low -engaged peers. To speculate, as one who is 111 familiar with the personalities and involvement of these 7 users in the institutional environment, I believe it is due to personal characteristics. All of these users are also those that are highly engaged in the institutional community, often accessing opportunities such as attendance at campus events, participation on college committees, and initiating interaction with colleagues. What might account for the low Socialwall participation? Some research on habit formation suggests that it may take closer to two months to impact behavior patterns (Lally et al., 2009), which means that the one-month intervention timeframe may not have been sufficient to build participation thr ough repeated use patterns. Other research related to contributions to online communities has indicate d similar patterns of passive and active use (Nonnecke & Preece, 2001), and has documented use that remains low across the course of a one -month interven tion period even when graduated incentive mechanisms are employed (Hummel et al., 2005). This again suggests that the timeframe may fall short of the ideal and also that it may be necessary to explore technology use and acceptance models to explore other factors for low participation. Next, I discuss how participants made use of the features of the space. Anticipated feature use. Approximately two -thirds of participants (n=15) interacted in the Socialwall by making posts and commenting on peer posts in the space. In Chapter II, I discussed the features of the Socialwall that might support socialization. I mentioned participants could take ownership of the experience and support access to information by using the Social wall to crowdsource questions, to drive interaction, and to mold what content was shared and discussed the space. Participants did utilize the Socialwall in this way, posting and leaving comments on peer posts , asking questions, 112 provid ing answers , sharing stories and perspective s, request ing information, and so on. Though 15 users made these types of typewritten actions in this space, only 7 visited the space regularly, and 9 participated but did not post at all. The scores and comments of those relatively disengaged users comprise a large subset in the data. This may help explain why this groupÕs perceptions of ownership on Likert -type and open -ended post -test questions failed to suggest that the Moodle Socialwall supported ownership. Also discussed in Chapter II , several studies have illustrated that social media provides an awareness of diverse perspectives and opportunities for peer modeling, which can motivate behavior offline (Beach & Doerr -Stevens, 2011; Greenhow, 2010; Robelia et al ., 2011). This was observed as participants shared assignment samples, discussed teaching tactics, and encouraged the attendance of their peers at upcoming college events. I also discussed how the Socialwall could support connectedness in Chapter II, by providing a venue fo r social support amongst peers and encouraging both bonding and bridging social capital. This also occurred in the space, with participants providing encouragement to peers for their ideas, acknowledging peers that they knew, and also commenting on how many of the people in the space they had not previously met. Unanticipated feature use. As noted in the analysis, only around a quarter of experimental participants (n=7) engaged with the Socialwall on a weekly, regular basis. When they did engage with typewritten actions, they were entered through the post and comment features on the Socialwall, posts in associated Moodle forums, and entries on other Moodle activities such as a glossary and a feedback activity. Participants did not make use of other features common to social media experiences such as the ability to ÒlikeÓ the posts and comments of peers, the ability to share hyperlinks and media, and the 113 ability to message directly to peers. Instead, they gravitated toward activities na tive to a traditional LMS experience in Moodle, such as posting in forums and participating in the Moodle feedback and glossary activities . Some of this may be due to the fact that I did not provide a great deal of overview or directions for use on the fe atures of the space and the benefit s of interacting in certain ways. Post -test responses also revealed that some participants were interested in aiding with the research or in being eligible for compensation by participating in the research, but that they did not have interest in the Socialwall or interaction with their peers that surpassed that level , which could have impacted participation . Because the Socialwall exists within Moodle, it is also quite likely that interactions were shaped by preconceived notions of how faculty should interact in the space from adjunctsÕ experiences teaching with traditional activities in Moodle (e.g., forums, feedback activities, etc.) . These activities are still available when the Socialwall format is applied , though th ey need to be added by a facilitator . In Chapter II , I wrote that participants could support connectedness and ownership through public identity development in the Socialwall by construct ing a profile with a picture , contact information, and indicating personal hobbies and interests. Though most participants did post a profile picture (n=19), and several participants commented that it helped them learn names or connect names with faces, no other user -generated features or f ields of the profile were used . I encouraged participants to add a profile picture in my introduction to the space and I modeled that behavior, but I did not encourage them to do anything else with the profile and I also did not model that behavior. 114 Socialwall interaction themes illustrating socialization . I listed and described the five primary themes that surfaced during Moodle Socialwall interactions (personal characteristics and initiative, information, connection with colleagues, technology, and challenges to onboarding) al ong with their sub -themes. Some of these overarching themes connected with the constructs of socialization. For instance, personal characteristics and initiative are a part of ownership, information also represents adequate information, and connection wi th colleagues represents connectedness. The sub -themes of these larger themes also illustrated connections with the constru cts of socialization . As an example , under the theme of personal characteristics and initiative, there were sub -themes that adjuncts should be active in connecting (connectedness), that they should pursue necessary information and resources (adequate information), and that they need to own their roles in the institution and department (ownership). The only surprise in the them es w as the prevalence of tec hnology in their comments and their expressed desire to pursue technology -aided solutions (Socialwall, Skype, video, etc.) to better engage the adjunct faculty. Challenges to onboarding were also a theme that was discussed, tho ugh this was anticipated due to themes in previous research with this population (Wicks, 2013). Ownership. Ownership, in the employed model for socialization, was represented through four Likert -type survey items regarding oneÕs likelihood to apply to the ir department if engaged in a job search, a sense of being valued in the work environment, a feeling of ownership of the department, and a sense of loyalty to the department. The need to feel valued was represented strongly in the data, as was the need to feel ownership, though it was stressed that adjuncts may need to take personal initiative to 115 build that feeling. Initiative is not represented in the current model. Participants also expressed a desire for the institution and administration to acknowled ge adjunctsÕ other commitments and roles as part of being valued. Items related to the likelihood to apply to the department in a job search and loyalty did not appear in the data. This may mean that this populationÕs satisfaction in the construct of own ership could be gauged more effectively by use of scale items that more closely align with the data presented here. In the review of literature presented in Chapter II, I discussed synergies between socialization and social media. First, I outlined how so cial media supports ownership . I suggested that it allows for participants to drive interactions that are customized to their interests, that it facilitates feedback from professional peers (Gruzd et al., 2012; Moran et al., 2011; Rowlands et al., 2011) , that it allows users to help generate content, and its support for peer modeling (Beach & Doerr -Stevens, 2011; Greenhow, 2010; Robelia et al., 2011) . Themes in the Socialwall interactions related to ownership shed light on my initial conceptualization . First, interactions primarily revolved around what they felt was necessary or what should be done to assist adjunct ownership of their role . One recurring theme throughout the Socialwall interactions was the need for the institution to show it value s adjun cts through inclusion and accessible opportunities. Including adjuncts in opportunities such as campus events and committee service through options that align with their schedules and other life demands, could help them feel more valued and access these opportunities to facilitate successful onboarding . This would help them feel like valued contributor s (ownership), gain the information and resources necessary to perform 116 in the role (adequ ate information), and develop ties to the college community (connectedness). In terms of actual interactions , I did observe that the mentioned affordances of social media were realized in the Socialwall space , though sometimes this was realized through traditional LMS activities and so metimes it was realized through wall posts, not with other features typical of social media (e.g., the ÒLikeÓ option for peer posts, sharing of hyperlinks and media, or personal messaging). This somewhat supported my hypothesis in the conceptual frame tha t the affordances of social media would readily aid socialization due to the participantsÕ exposure to a variety of others and information in the space , though it must be considered in the context of the small subset verbal contributors to these comments ( n=15). For instance, participants drove the conversation with their interests and assisted in generating content as illustrated in the following comment: ÒI would like to see a structured conversation (forum?) on the topic of best practices in online teac hing. What are the most effective practices that help students to be engaged (interested/participating/learning) in the online experience? What are the challenges to online learning and how do we overcome them?Ó (Stan, male, late -50Õs, adjunct for 33 yea rs) Participants also offered each other professional feedback and practiced peer modeling as illustrated in the following exchange: ÒI require every student to give an impromptu speech each week. I tie the topic to the Chapter we are covering that week. The speeches are short Ð usually one or two questions that they can answer quicklyÉHere are a few 117 examples from previous Weekly AgendasÓ (Jim, male, early -60Õs, adjunct for 5 years) ÒJim! I LOVE this! Very cool. How do you support the students with difficulty speaking in front of groups in this activity?Ó (Veronica, female, early -50Õs adjunct for 15 years) Information. Adequate information, in this model, was represented through four Likert -type survey items re garding the receipt of adequate information to serve, the access to adequate resources to serve, clear explanations of departmental service expectations, and the participantÕs understanding of expectations for continued employment at the institution. The needs for ready access to up -to-date information and resources were strongly represented in the data. However, participants were less focused on service expectations and expectations for continued employment and more focused on information and feedback th at would help them be more effective educators (e.g., implementing the curriculum in a consistent manner and fostering student success). Like the data for ownership, data for the construct of adequate information suggests that adjunctsÕ satisfaction in th is construct may be more effectively gauged with survey items that aligned more closely with their expressed concerns and interests. In Chapter II, I also detailed how social media supports access to information. I noted its ready accessibility , its suppo rt for quick sharing and crowdsourcing, its capacity for targeted communications, and how it promotes exposure to diverse perspectives (Greenhow, 2010; Robelia et al., 2011) . There was a theme of interaction related to the topic of adequate information in addition to sub -themes in each area . Again, these comments were usually phrased in terms of things that institutions and administrations 118 should be doing. For example, common strands ac ross all themes were that we as administrators should provide necessary and relevant information and feed back to adjuncts. In terms of actual interactions, I again observed that the indicated affordances of social media were utilized in the Socialwall spac e, which again somewhat supported the assumptions of my conceptual frame. However, I also echo here that this must be considered in the context of the very small contributing sample. Interactions represented both targeted communications and exposure to d iverse perspectives, as shown in this exchange: ÒI would like to see if this could be a useful tool for all. If this could be moved to the Moodle shell, everyone could participate. (Jim, male, early -60Õs, adjunct for 5 years) ÒAlthough I agree with Jim th at everyone should be able to use this social media outlet, I do appreciate a smaller, more familiar group, and departments would achieve that. I would NOT recommend dividing by adjunct/full -time faculty. Is there a way to do both? (Veronica, female, early-50Õs, adjunct for 15 years Additionally, interactions demonstrated crowdsourcing, as indicated in this comment: ÒThe new version of Moodle is too slow and seems to have a number of stop script messages, although the last one may be my computer. I am cu rious if others have this problemÓ (Kenny, male, mid -60Õs, adjunct for 6 years). In terms of accessibility, participants commented on the easy access of the space in their open -ended survey responses. 119 Connectedness. Connectedness, in this model, was represented through four Likert -type survey items regarding a feeling of connection to others in the work environment, opportunities for socialization with colleagues, opportunities for formal interactions with colleagues during work, and the sense that oneÕs colleagues are also considered to be friends. Here, data acknowledged adjunctsÕ need for connection with others in the work environment, socialization with colleagues, and opportunities to interact during work (specif ically with others in the department). Data did not suggest that this population had a strong need or desire for friendship with colleagues, though there was a focus on accessible inclusion, such as at college events and committee meetings. In alignment with findings related to the other constructs of socialization, the data here suggest that the construct of connectedness may be more effectively assessed through a slight adjustment of survey items. Lastly, in Chapter II, I discussed how socia l media supp orts connectedness. I mention ed its capacity for ongoing asynchronous participation, including both personal sharing and support amongst peers, how it promotes collaboration along with promoting offline engagement and interactions (Greenhow & Burton, 2011 ; Junco, 2012 Steinfield et al., 2008) , and finally how it promot es both bonding and bridging social capital (Burke et al., 2010; Ellison et al., 2007; Ellison et al., 2014; Valenzuela et al., 2009) . Here, as for adequate information, connectedness appeared as a major conversational theme and a sub -theme threaded through the other interactions. Connection with colleagues was one of the primary topics discussed and sub -themes on this topic varies from the need to foster connection with, and provide s upport to, adjuncts along with providing opportunities for connection that acknowledge adjunctsÕ other 120 commitments. As in the other areas discussed, these comments were framed in terms of what adjuncts and administrators should be doing to facilitate conn ection. Also similar to the other a reas discussed, Socialwall interactions demonstrated that participants were making use of the affordances of social media to meet their socialization needs, which again somewhat supported the conceptual frame , though sti ll considered in the context of the small sample . For example, interactions included sharing along with support and bonding as illustrated here: ÒIÕve always had a vivid imagination. Using creativity to express my personality is part of my life. I enjo y a good art project or a good challenge to push my creative thinkingÉWanna take a Creativity [hyperlink] test?Ó (Jennifer, female, early -30Õs, adjunct for 5 years) ÒThank you for sharing this! IÕm not sure what it says about me (56%) but it doesnÕt soun d goodÓ (Jane, female, late -40Õs, adjunct for 8 years) There were also moments at which online interactions encouraged offline interactions, as shown by this comment, ÒThe next [campus omitted] Education on the Edge lunchtime gathering will be on Wednesday, March 4 in [room omitted]. Lunch will be fixed by [name omitted]. All faculty are welcome, including those who havenÕt ever come to an E -on-the -E discussion before.Ó (Jim, male, early -60Õs, adjunct for 5 years) Finally , here is an example of h ow Socialwall interactions helped promote bridging social capital: ÒThis seems crazyÉI left FT status at [institution name] just two years ago and I 121 only recognize 15% of the names on this adjunct pilot.Ó (Robin, female, late -50Õs, adjunct for 2 years) Summary. In general, the themes of interaction in terms of topics discussed and in terms of the types of interactions demonstrated in the space were as expected for Socialwall participants. Though there were some interesting additional topics shared by small pods of users related to the need for institutional focus on product consistency and quality along with student success, the primary surprises came, not in the form of conversational themes, but in terms of software use patterns. Here, I summarize fe ature use and the content of interactions. Feature use. As previously discussed, use was generally quite low, which may be due to the design of the intervention and the lack of a clear purpose and instructions for use for this audience. It also may be du e to the brief timeframe that users had to incorporate the tool into a regular routine, which suggests that future intervention design will want to consider an extended period for use and data collection. However, as previously noted, some research sugges ts that this participation is consistent with that in other studies related to online community contributions (Schellens & Valcke, 2005). It was surprising that the users who did engage regularly by visiting the space weekly made even less use of more act ive forms of communication available in this social media space, such as posts and comments, than did their peers with lower weekly engagement. However, Òlurking,Ó or the tendency for online community members to engage in less active forms of participatio n such as spending time visiting pages and reading peer posts, is a common and widely -studied phenomenon in the study of online communities that may still reap benefits for those participating in these ways (Arnold & Paulus, 2010; 122 Beaudoin, 2002; Preece, N onnecke, & Andrews, 2004; Rau, Gao, & Ding, 2008). Likewise, it was surprising that this audience made very low use of some of the features common to social media outside of posts and comments, which include the ability to ÒlikeÓ peer posts, to share link s and media, and to message with users. This may be due to the lack of direction that participants were given in the space explaining what features the software offered and how to use those. More purposeful intervention design including training on socia l media features and their benefits may have impact on use . It may also be that the brief timeframe did not allow adequate time for them to integrate it into their routines. Content of interactions. According to this population, there were five major areas of interest in the Socialwall. These were: (1) personal characteristics and initiative, (2) information, (3) connection with colleagues, (4) technology, and (5) challenges to onboarding. A great d eal of conversation was focused on things that the administration might do to assist the socialization of adjunct faculty, which is not surprising in the context of the informed consent form that participants read and the pre -test they completed prior to i nteracting in the Socialwall space. Though experimental group participants exhibited many of the benefits afforded by social media in their interactions (e.g., peer support, bonding and bridging social capital, exposure to diverse perspectives, driving th e conversation based on their interests, etc.), they did not always realize these benefits by using features unique to social media such as the ability to ÒlikeÓ peer posts, to share hyperlinks and media, and to message with others individually. In terms o f how these constructs of socialization were defined, conversational themes illustrated that the Likert -type scale items for this model of socialization may 123 need to be reconfigured to more effectively measure the components of socialization that are import ant for this population. For instance, in terms of ownership, t he need to feel valued and the need to feel ownership were present in the qualitative data, but there were not clear links to the items on job search and loyalty. Personal initiative and inst itiutional acknowledgement of adjunctsÕ other commitments were present in the data, so those items may need to be explored. In terms of information, the Likert -type items focused on ready access to up -to-date information and resources were echoed in the d ata, though the service expectations and expectations for continued employment seemed absent. Instead, participants mentioned information and feedback that would help them be more effective educators in their responses. Lastly, in terms of connectedness, adjunctsÕ need for connection with others in the work environment, socialization with colleagues, and opportunities to interact during work (with others in the department) were apparent . However, d ata did not suggest the Likert -type item on friendship wi th colleagues was relevant. Instead, ba sed on the data, it may be beneficial to explore a scale item related to opportunities for accessible inclusion in college events and proceedings . Research Question 3 The third research question (RQ3: What benefits, if any, did the experimental group perceive that the intervention provided to support their socialization? ) was analyzed through Likert -type survey data and through thematic content analysis of open -ended survey responses and of Socialwall comments and pos ts that were directly related to the Socialwall tool and/or experience. Here, I provide more information about these trends along with commentary. 124 Likert -type data. Likert -type questions related to perceptions of how the intervention supported the three constructs of socialization revealed very low means for ownership. Approximately half of the small subset of experimental group participants neither agreed nor disagreed that the intervention supported their sense of ownership, while nearly a third disagr eed or strongly disagreed. As stated previously, the control group displayed low means for ownership, decreasing in a statistically -significant way by the end of the study. That may be due to the fact that the intervention was withheld from the control g roup, some of who really wanted to trial the Socialwall software. I also postulated that the adjunct group as a whole may struggle with the concept of ownership given their unique appointment model as Òat will, at needÓ employees . Results were more posi tively skewed for information and connectedness, where about a third of experimental group participants neither agreed nor disagreed that the intervention supported either of these constructs, though over 50% of this small sample agreed or strongly agreed that it did. These scores make sense in the context of the themes that emerged in both the Socialwall (see Table 5) and in the participant post -test comments about the intervention (s ee Table 4 ). In both cases, the ideas of access to adequate information and connectedness seemed to be frequently -discussed topics and readily recognized needs for intervention participants. Post -test survey responses and constructs of socialization. In relation to the open -ended survey items, responses could be categorized i nto five major themes describing how the verbal subset of experimental participants (n=15) perceived that the intervention supported the constructs of socialization: none, collaboration, connection, information, diverse viewpoints, tech tools. Similar to the interaction themes from the 125 Socialwall space, some of these post -test survey response themes were also related to the three constructs of socialization, while many of the sub -themes were clearly linked to those constructs . Noteworthy points here were that a sizable portion of experimental group participants largely failed to perceive any features of the Socialwall that supported their senses of ownership, but that a subset of them also readily recognized a failure to take ownership of their participation in this experiment with the Socialwall. This echoes the general sentiments of low perceived ownership indicated by control group participants on the quantitative measures and begs the question raised above of whether or not all adjunct faculty collectively struggle with a sense of ownership due to their contingent appointments and other life commitments. It also aligns with the theme of ownership communicated in the Social wall space indicted in results for RQ2, in which participants seemed to be suggesting that their adjunct peers needed to take more ownership in their socialization. Socialwall comments about the Socialwall experience. Additionally, comments and posts abo ut the Socialwall made in the Socialwall space were analyzed thematically, resulting in three different themes (purpose, positivity/potential, and functionality) that described how users viewed the space immediately and across time. Though none, as in no noted gains (n=15), was a theme of user perceptions on the combined post -test questions discussed above, the Socialwall data was quite positive, illustrating positivity and potential ( 13 comments from 12 users ). So, while participants seemed less able to articulate benefits of the Socialwall for socialization on survey questions, they were positive about the experience and the potential of the Socialwall for socialization in their comments in that social media space. This may again be due to the fact that the one - 126 month timeframe was not enough time for the users to report immediate benefits of use, though it may have been enough to evaluate potential benefits in future use models. This data also suggested that, in future research with this tool, it may be useful to clarify the purpose for using the tool (n= 8) and to address tool functionality (n= 9) to media te some of the uncertainty expressed by participants. Connecting the Research Findings In considering the data from all three of the research questions , each provides a bit of context to more fully understand the others. For instance, in RQ 1, I found that there was no stati stically-significant difference in the experimental groupÕs mean socialization scores across time, or in comparison to their control group peers. This is better understood when considering the quantitative data collected for RQ2, which revealed low and inconsistent use of the Socialwal l space along with very limited use of the social media fe atures available in the space. Additionally, when considering the data from RQ3, its apparent that a small subset of participants perceived support in the experience for their access to adequate in formation and connectedness, though many of this subset did not believe that the experience supported their sense of ownership. This was echoed in the qualitative data, in which users named benefits for the former, but over half named no benefits for the latter. It was evident through Socialwall activity that users were not making use of some of the features designed for ownership, like the option for profile development, while they were making use of others, like the ability to drive the conversation bas ed on their interests. This suggests that it may be beneficial to provide users with an overview of Socialwall features and their benefits in the design of an 127 intervention and also to increase the duration of an intervention period to allow them enough ti me to adopt the tool into their daily routines. However, t he benefits perceived for information and connectedness were represented inside the limited interactions that occurred in the Socialwall space. Users made use of features designed to support these capacities such as making posts and comments to share information, to support one another, to ask questions, to encourage offline participation with their peers, and to explore diverse perspectives. This is underscored by the data collected in RQ3 from in teractions in the Socialwall that referred to the Socialwall tool and experience. When users spoke about the Socialwall experience, there was some lack of clarity on its purpose and utility, but there were a higher number of positive comments from this sm all group that also discussed its potential. The usersÕ perception of value in the tool coupled with its low cost of implementation and maintenance raise questions as to its value for continued exploration as a tool for adjunct faculty socialization , desp ite the fact that quantitative evidence did not support its utility for this purpose . In terms of this model of socialization and its applicability to this population, the interaction themes provide points for consideration. The first point to consider is whether or not this model for socialization that was formulated in a 4 -year setting is truly applicable for these community college adjunct faculty members. The typewritten actions analyzed in RQ2 helped define adjunct needs and potential benefits of the space. The content of the interactions were focused on ( 1) personal characteristics and initiative, (2) information, (3) connection with colleagues, (4) technology, and (5) challenges to onboarding. User interactions helped identify potential changes for the Likert -scale items 128 utilized to measure the constructs of socialization for this audience. For instance, likelihood to apply to department in a job search (ownership) , loyalty to the department (ownership) , service expectations (information) , expec tations for continued employment (information) , and friendship with colleagues (connectedness) were not items of note in conversational themes . Thus, they may not be as critical in gauging participant satisfaction with these constructs for this population , though they were found to be important in another faculty population (Schrodt et al., 2003) . Additionally, items such as personal initiative (ownership) , institutional acknowledgement of other commitments (ownership) , information and feedback to help ad juncts be more effective educators (information) , work with department colleagues (connectedness) , and accessible opportunities for inclusion in college events (connectedness) did surface as critical themes in the Socialwall interactions that were relevant for this group . Thus, it may be useful to explore adding these as scale items and computing reliability for the constructs with these additions among this adjunct faculty population . Limitations Limitations in design. The pre -test post -test design offers much utility in terms of group comparison; however, it is limited in the threats to external validity inherent in this design. Primarily, it is quite limited in its scope. Results are not generalizable to any audience, setting, or time period beyo nd the present study given the unique population, context, and socio -political period in which this research is being conducted. That is not a major concern for this study, as the goals were to explore a tool that might impact the socialization experience of this unique group and also to serve as a conceptual model that could be trialed in other contexts. Halo effect (Thorndike, 1920) may be a nother 129 limitation , related to my role as a discussion facilitator in this design . In ThorndikeÕs conceptualizatio n of the halo effect, participant scores or comments may be positively skewed due to positive feelings about a person. This is a realistic threat, as one of the experimental group post -test participants reported that the tangible benefit of participation was having an opportunity to aid me personally in my quest for a doctoral degree. Limitations in sampling. Unique to this particular study was a limitation in sampling. Chiefly, the sample size (N=77) was very small when compared with the a priori power analysis recommending an ideal sample of 222. In a post hoc power analysis, inputting the participant numbers for the experimental and control group indicated that the power for the study was quite small, just .23. This means that the quantitative result s of this study may not hold true for the larger population. For instance, the qualitative theme codes were derived from a very small, verbal subset of participants. This may mean that results are skewed to match those participantsÕ experiences . Many di d not participate, which could suggest that the Socialwall is not actually a solution for all those in the target population, though qualitative themes might suggest otherwise. Another potential limitation in sampling may be a volunteer bias in which part icipants with certain characteristics or predispositions are overrepresented in the sample. For instance, this sample was skewed with a higher percentage of female participants than is represented in the broader population. Was the social media intervent ion more appealing to female participants due to the societal role of women in the United States as social gatekeepers? Additionally, it was not possible to isolate research participants from one another during the course of the research. If participants 130 from either group have discussed the instruments and the intervention with one another, it could contaminate results. Limitations in p rocedures. Perhaps most prominent in limitations were some related to procedures. First, the start of the research was delayed. The intent was to launch the research with the launch of the new semester in winter. Due to issues with the timetable for acquiring the code for this software plug -in, the research did not begin until mid-semester, when exams and the traditional spring break week were occurring on campus. The timing may account for some of the attrition and the low activity from many experimental group participants. Additionally, the one month timeframe for the research was quite brief. One study related to ha bit formation suggests that it may, on average, take over two months to impact behavior patterns (Lally et al. , 2009). So, the time allotted to the study may not have served as a reasonable enough timeframe for participants to adopt the Moodle Socialwall into their regular routines and to perceive its potential benefits. The data analysis techniques may also have presented a limitation. Though paired samples t -tests were selected to compare the outcome means for one group across time and independent sam ples t -tests were selected to compare the outcome means for two groups on the post -test, I also extracted the only -adjunct group from the experimental sample to compare their scores due to the fact that those holding multiple roles at the college may have more opportunities for socialization. A dissertation committee member indicated that the use of analysis of covariance (ANCOVA) might have presented a more fitting option for analysis. I recognize that ANCOVA m ight have better addressed the comparison of means on more than two independent groups and might have increased the sensitivity of my test , when compared to t -tests, by controlling 131 for a potential confounding, continuous variable such as employee group af filiation (Pallant, 2013). This will be cons idered in future statistical analyses with this data set. As a final consideration in this arena, given that the procedural measures relied mainly on self-reported data, it should be noted that the user reports were subject to bias and selective memory, a mong other things. Implications and Recommendations The goals of this study were twofold: (1) to explore the efficacy of social media as a low -cost solution for the organizational socialization of adjunct faculty members, specifically in the form of the n ew Socialwall format plug -in for the Moodle LMS, and (2) to understand how a specific population of adjunct faculty members interact and perceive that interaction with peers in the Moodle Socialwall impacts their organizational socialization. Sizable oppo rtunities may exist for low -cost, low -maintenance connection and development that have yet to be realized in practice or in the literature in these areas. In the tradition of scholarly work, this study has generated many more questions than it has answere d. One could likely draw no definitive answers about the potential for social media in adjunct faculty development from the data collected here. That said, the overall insights were twofold: (1) a small subgroup of participants felt that social media ha d the capacity to facilitate adjunct facultyÕs socialization, particularly by providing access to information and by supporting connection with peers ; and (2) there are factors impacting socialization for this population that warrant further study (personal initiative, institutional acknowledgement of commitments, information and feedback related to instructional efficacy, formal departmental interactions, and accessible opportunities for college inclusion). Moreover , this study suggests further research areas and practical 132 implications related to social media use and design for educator socialization and the use of social media in c ommunity colleges environments. To aid research and practice, I offer several recommendations throughout this next section. Implications and recommendations for practice. Participant sÕ Socialwall interactions and survey responses indicated positivity and perceived value in Socialwall use for the constructs of adequate information and connection. Base d on this feedback , there is merit in the continued exploration of social media like the Socialwall for the practice of onboarding adjuncts. Social media provides a low -cost, low -maintenance option to facilitate the socialization of adjunct faculty. In al ignment with the findings of this research , I would make two practical recommendations to administration and adjunct faculty developers looking to implement s oftware like the Socialwall for this purpose. First, practitioners may want to consider a semester -long trial of a social m edia environment like the Moodle Socialwall, or a trial sp anning a full academic year. This would allow them to more effectively measure its impact . With this timespan, practitioners could help ensure that sufficient time is allotted for participants to adopt the tool into their routine and to perceive its impacts. As mentioned previously, further reading on habit formation suggests that it may, on average, take 66 days to impact behavior patterns (Lally et al., 2009). This means that more than two months may be necessary to impact behavior, which lends itself to the recommendation for the increased trial. Second, intervention design should be carefully considered. The structure of this intervention was purposefully open -ended, with little introduction, scaffolding, or instruction to allow for organic use patterns to develop amongst participants. However, i t 133 may be that more focused efforts are necessary to encourage consistent participation. Previous research regarding participation in online communities has suggested that tiered incentive structures may scaffold more consistent contributions (Hummel et al., 2005). However, intervention design could include a variety of other elements such as initial training, scaffolding or modeling patterns of interaction, determining relevant content, and more. Based on the fact that these participants made little use of features common to social media and gravitated toward traditional LMS features, I would recomme nd that the initial training piece with inter vention participants include an overview of the software features, their potential uses, and potential gains . Because there was confusion about the purpose of the software in this study, I would also recommend that practitioners consider presenting a clear purpose for social media use to see if this encourages higher engagement. Lastly, participants often made suggestions related to technology -facilitated socialization options to increase accessibility. Howeve r, they also voiced a need for technical support in using technology. For these reasons, I would recommend emphasizing technical support contacts when implementing a social media tool for socialization. Implications and recommendations for research. As stated above, participant Socialwall interactions and survey responses indicated positivity and perceived value in Socialwall use for the constructs of adequate information and connection. Based on this feedback, there is merit in the continued explorati on of social media like the Socialwall for the practice of onboarding adjuncts. Though I previously spoke of implications for practice, here I speak of implications for research. This conceptual model of onboarding geographically -dispersed and varied aud iences with a low -cost, low -maintenance solution 134 has value for a variety of audiences, such as adjunct faculty, online faculty, and the commuter student and faculty audiences common to community college settings. For those interested in examining social m edia tools, like the Socialwall, for the socialization of similar adjunct faculty, I offer several recommendations for the design of research and research questions. A critical point to consider in terms of research is how to increase the sample size for m ore power in results. There may be institutional tactics that would encourage participation. For instance, I might suggest that instructional administrators become involved in the effort, encouraging participation from their adjuncts and acting in somewh at of a facilitator role as I have in this study. This may encourage participation because it would provide a direct line of communication for adjuncts with their supervisors, and some members of this population expressed a desire for more feedback from t heir instructional administrators in comments in this study and in previous research (Wicks, 2013). Another method of increasing sample size discussed in the proposal phase for this research was the idea of also trialing the tool at another institution wi th need for support for their adjuncts. That option was not pursued in this research for several reasons; first, because the goal of this study was to address some of the unique needs for connection expressed by this specific population of adjuncts and, s econd, because of the logistics and timeframe limitations. It will take some time to connect with another institution using Moodle, to garner interest and support from their instructional administration and technical teams, to implement the Socialwall plu g-in on their installation of Moodle, and to negotiate shared procedures for the study. Lastly, one option for increasing sample size may be to include full -time faculty in the research. As 135 previously discussed, many adjuncts participating in the Socialw all expressed interest in gaining more connection with full -time faculty and several suggested that the Socialwall might be an excellent tool for communication within academic departments. That was not considered in this study due to the goal of exploring solutions for the unique needs of adjuncts and the very different appointment model in which full -time faculty serve. However, the model for socialization used within this study was formed in the context of new full -time faculty in the communications dis cipline (Schrot, et al., 2003), so it warrants further consideration. Second, more research may be necessary on technology use and acceptance frameworks to sculpt the conditions in which users may be more willing and more likely to participate. Previous research with this population (Wicks, 2013) has indicated that the Unified Theory of Acceptance and Use of Technology (UTAUT) (Venkatesh, Morris, Davis, & Davis, 2003) may not be a fit for this population in terms of understanding their social media use . UTAUT proposes that users will be more likely to use a technology if they perceive that: (1) using the tool will enhance their performance, (2) the tool will be easy to use, (2) the organizational structure will provide support in using the tool, and (4) others think they should use the tool (Venkatesh et al., 2003). These principles did not hold true for this populationÕs use of social media in that research (Wicks, 2013). For example, though there was a moderate correlation indicated between this popul ationÕs perceptions that social media would enhance their performance and their actual use of social media, correlations related to the other indicated criteria were negligible. However , many other such frameworks exist that may inform implementation and support. Intervention design could also be considered. The structure of this intervention 136 was purposefully open -ended to allow for organic use patterns to develop amongst participants. It may be more that focused efforts are necessary to encourage consi stent participation. Previous research regarding participation in online communities has suggested that tiered incentive structure s may scaffold more consistent contribution s (Hummel et al., 2005). Implications and recommendations for theory. One question that is generated from this study is how appropriate this model of socialization is for an adjunct faculty population. As mentioned in both the methodology and the results, CronbachÕs Alpha scores were calculated to examine the cohesiveness o f the socialization scale employed in this study. Each of the alpha scores exceeded the cutoff score, confirming the reliability of the scale and the interrelation of items on it. However, as stated, this scale was originally developed by Schrodt et al. (2003) in a study of new communication faculty, most of whom were employed in 4 -year institutions, and many of who were in tenure track positions. Given the fact that the results of this intervention study did not quantitatively support positive changes i n perceptions of this model for organizational socialization and its constructs, it may be necessary to ask and explore how closely the scale items align with the issues of concern for onboarding specific to th is adjunct faculty population. For instance, o n open -ended pre - and post -test comments on additional aspects of socialization that may be important to participants, several topics were mentioned that might not b e accounted for in this model. Outside of themes directly related to ownership, adequate i nformation, and connectedness, Socialwall interactions identified themes related to technology and challenges to onboarding. Additionally, user 137 interactions helped identify potential changes for the Likert -scale items utilized to measure the constructs of socialization for this audience. To provide several examples, likelihood to apply to department in a job search (ownership), loyalty to the department (ownership), service expectations (information), expectations for continued employment (information), a nd friendship with colleagues (connectedness) were not topics that emerged in conversational themes. Thus, they may not be as critical in gauging participant satisfaction with these constructs for this population. Conversely, items such as personal initi ative (ownership), institutional acknowledgement of other commitments (ownership), information and feedback to help adjuncts be more effective educators (information), work with department colleagues (connectedness), and accessible opportunities for inclus ion in college events (connectedness) did surface as critical themes. Thus, it may be useful to explore adding these as scale items and computing reliability for the constructs with these additions. Likewise, i t may be that the framework for faculty organ izational socialization presented by Schrodt et al. (2003) is not well -suited for addressing the problems associated with adjunct faculty onboarding in a community college . As stated, it was developed within a four -year setting in which many new and conti ngent faculty are seeking long -term, full -time appointments. It may be that the contingent nature of adjunct faculty appointments does not attract an audience with a quest for connection with colleagues and an ownership of the adjunct role because many co mmunity college adjuncts are not seeking long -term, full -time appointments. Additionally, i f this problem were re -framed with a different theoretical perspective , it might prompt vastly different solutions, some of which may more closely address adjunct n eeds for connection and 138 development. For instance, communities of practice (CoP ) (Lave & Wegner, 1991) and virtual communities of practice ( VCoP) ( Dub”, Bourhis, & Jacob, 2005) are a popular concept and model in education. According to Wegner (2007), three elements are common to communities of practice: (1) a shared domain of interest such as hobby, profession, or discipline; (2) community engagement in terms of conversations, shared information, and relationship building; and (3) practice, meaning mem bers are those who do the things they discuss and share their experiences and tools for group development. If adjunct faculty onboarding were viewed from this perspective, a virtual community of practice might include conversational prompts that were more closely related to teaching practice, which could still build the information -sharing, connection, and ownership desired. Another potential model out of the many possible models might be the soci al constructionist framework presented by Vygotsky (1978). The basic tenets of VygotskyÕs views on social constructivism are: (1) learning and development are social and collaborative activities, (2) there is a zone of proximal development (ZPD) for learners in which they can reach a milestone with assistance, or scaffolding, (3) learning activities should be realistic, and (4) learning occurs in both formal and informal settings and we must capitalize on the interactions between these settings. If adjunct faculty onboarding were viewed from this perceptive, a vi rtual social constructivist community might include conversational prompts that are related to critical job tasks with support from full -time faculty and administrators. This could scaffold the learning and development of participants. In fact, given tha t the experimental participants reported a desire for information and feedback related to instructional efficacy and for formal departmental interactions, which were not mentioned in the framework employed here, it 139 may be that a community of practice or a social constructivist frame more closely meets this populationÕs needs. Conclusion In summary, the organizational socialization of adjunct faculty is both a critical and timely issue that warrants further consideration in both research and practice. Current approaches fall short of adjunct needs (Gillespie & Robertson, 2010; Langen, 2011; Rouche et al., 1996; Shannon, 20007) in terms of content and scheduling. A model of socialization like that proposed by Schrodt and colleagues (2003) provides a useful starting point in exploring adjunct facultyÕs satisfaction with the socialization experience, as this population of adjuncts has expressed notable opportunities related to both their perceptions of ownership of the role and connectedness to the institution, which seem nearly universal concerns in the adjunct -related literature (Finucane & Algren, 1997; Meixner et al., 2010; Wicks, 2013). Social media provides documented opportunities for information sharing (Gruzd et al ., 2012; Rowlands et al., 2011), identity development (Byrne, 2007; Davis et al., 2012), and the building of social capital (Burke et al., 2010; Ellison et al., 2007; Elliston et al., 2014) among other things, which could help meet adjunct needs. Social m edia also provides a low -cost, low -maintenance solution for socialization, given its potential for crowdsourcing, and the fact that many academics are naturally opting into social media due to its capacities for facilitating professional connections, colla boration, and dissemination (Moran et al., 2011; Rowlands et al., 2011). Additionally, it provides a platform for administration and full -time faculty to hear the needs and concerns of adjunct faculty and act responsively. 140 However, in implementing social media solution s for adjunct faculty socialization , careful consideration should be directed at implementation procedures and timing to help sculpt useful and enduring participation with the tool from its user community. In future study, attention should b e directed to frameworks that better suit the needs expressed by community college adjunct populations , technology adoption models, and elements of research design. Specifically, it may be worthwhile to ask how relevant the model of socialization employed here is in the context of adjunct faculty. Noteworthy topics raised by this population of adjunct faculty included factors such as personal initiative, instructional content focused on instructional efficacy, flexible and accessible options for college i nclusion, and so on. There may be conceptualizations of socialization that already incorporate these factors and are, thus, more relevant to the topic of onboarding adjuncts. There also may be other frameworks, such as communities of practice or social c onstructivism, that better frame and meet these expressed needs. Additionally, future study on social media intervention s will want to consider models of technology acceptance and adoption that may undergird implementation tactics when employing social media in these means. Finally, in terms of research design, it will be critical to access large r samples to more accurately gauge statistical significance, to introduce the tool and its features in more purposeful with adjunct faculty to promote engagement and full feature use, and to explore extended timeframes for intervention testing that allow participants to incorporate t he tool into their ongoing routines . 141 APPENDICES 142 APPENDIX A Call for Participation Hello, [Name]. Would you be willing to take part in a compensated research project informing adjunct initiatives and Moodle development at MMCC? As a member of our active adjunct faculty pool who makes use of learning technologies in your practice, you are best equipped to assist in exploring these issues. I hope to begin this study within the next 2 -3 weeks, so your prompt response is esse ntial in our planning. There are two options for paid participation: ¥ Option1: Complete two brief, online surveys about your experience as an adjunct faculty member at MMCC and be compensated for each survey. ¥ Option 2: Complete two brief, online surveys a nd be compensated as described above. Also, help us test the new Moodle eCommunity plug -in, which helps a Moodle course act like a social networking space. eCommunity participants will be included in a drawing of several sizeable gift cards for this extr a effort. You may recall my previous survey of adjuncts at MMCC. As a fellow adjunct faculty member and the Director of Distance Learning, I'm very interested in continuing these efforts. And, as a current Ph.D. candidate at MSU, this research will assi st me in completing the dissertation requirement of my doctoral degree. Please review the form below prior to Sunday, March 1, 2015 at midnight EST to learn more and to offer your consent if you're willing to participate. I thank you in advance for your timely response! Information and Consent Form Link: [Survey Link] Or copy and paste the URL below into your internet browser: [Direct URL] Regards , Jessica Wicks Director of Distance Learning, Multimedia, & Instructional Design (DLmid) Mid Michigan Community College 989.317.4601 or x101 jwicks@midmich.edu or wicksjes@msu.edu Follow the link below if you wish to opt out of future emails related to this research project: [Link to Unsubscribe] 143 APPENDIX B Informed Consent Thank you for following this link! You are being asked to participate in a research project at MMCC related to the topics of ongoing adjunct faculty socialization and the Moodle eCommunity plug -in, which helps a Moodle course act as a social networking sp ace. Purpose - Your participation may help inform initiatives for adjunct faculty support at MMCC and, depending on your level of involvement; your participation may also help us test a newly installed Moodle plug -in called the Moodle eCommunity Format. Procedures - Participants will be randomly assigned to one of two groups: ¥ Group 1 - You are asked to complete two brief, online surveys on your experience as an adjunct at MMCC across the course of one month. ¥ Group 2 - You are enrolled in an adjunct eCommunity within Moodle and are asked to interact in it throughout one month. You are also asked to complete two brief, online surveys on your experience as an adjunct at MMCC across the course of one month. Risks & Benefits - This study holds no notab le risks to participants. However, benefits include a potential opportunity to engage with colleagues. Compensation - All participants are eligible for $5 per each completed survey, which will be disbursed within 48 hours of the survey close dates. eCo mmunity participants are also eligible for a drawing of 4 -$100 gift cards that will be awarded intermittently during the course of the month of eCommunity participation. Drawing winners will be determined through the use of a random number generator. Confidentiality - Though the survey responses and interactions that take place within the eCommunity space are not anonymous, records will be kept confidential to the researcher. When reporting the outcomes of this research, participant names or other pie ces of information that may individually identify participants will not be used. Voluntary Nature of the Study - Participation is completely voluntary. Your decision to participate or not will not affect your current or future relationship with MMCC. You may refuse to participate in certain procedures or to answer certain questions or you may discontinue your participation at any time without consequence. Contacts & Questions - If you have concerns or questions about this study, please contact the res earchers: Primary Contact : Dr. Christine Greenhow Assistant Professor, Counseling, Educational Psychology, & Special Education Michigan State University College of Education 620 Farm Lane, Erickson 513F East Lansing, MI 48824 -1034 greenhow@msu.edu 517.432.0425 Secondary Contact : Jessica Wicks Director of Distance Learning, Multimedia, & Instructional Design (DLmid) Mid Michigan Community College 144 2600 S. Summerton Road, Doan 108 Mt. Pleasant, MI 48858 jwicks@midmich.edu 989.317.4601 or x101 If you have questions or concerns about your role and rights as a research participant, would like to obtain information or offer input, or would like to register a complaint about this study, you may contact, anonymously if you wish, the Michigan State Universit y's Human Research Protection Program at 517.355.2180, Fax 517.432.4502, or email irb@msu.edu or regular mail at 207 Olds Hall, MSU, East Lansing, MI 48824. Consent - By clicking below, you indicate that you have read this form, you understand its terms, and you voluntarily agree to participate in this research project. ! I consent to be involved in this research study. Payment Arrangements - Any participants teaching at the MMCC Harrison or MMCC Mt. Pleasant campuses will receive the promised compensation via interoffice mail. If, however, you only teach at an offsite or online location, please indicate preferred payment arrangements below: ! I can make arrangements to visit the MMCC Harrison or MMCC Mt. Pleasant campuses to collect payment for my participation. ! I ask that payment for my participation be mailed to the home address that I have on file with the College. ! Other (Please indicate). 145 APPENDIX C Notice of Group Assignment - Control Hello, [Name]. Thank you for consenting to be involved in research on the MMCC adjunct experience and the new Moodle Socialwall format plug -in. (Though I previously referred to this new Moodle format as the eCommunity format, representatives from Moodle HQ suggested the name change and has recently been adopted by the developers.) Group Assignment You have been randomly assigned to Group 1, surveys only . I know some participants requested a specific group, which is unfortunately not an option. A randomized design allows me to capture comprehensive feedback from a wider audience of users that more accurately reflects the larger stakeholder group. Your continued participation is, of course, completely optional. Those who were very interested in participating in the Moodle Socialwall pilot group will still have the opportunity to do so in the future once this one -month timeframe for the research has passed. What You're Being Asked to Do I'm asking you to complete a survey this week and a follow -up survey in one month . Both are brief and I offer $5 each for their submission. Next Steps (Action Necessary By Saturday, March 7 at midnight EST) Please take the following steps to signify continued involvement in this research: Complete the survey at the link below by Saturday, March 7 at midnight EST Survey Link: [Survey Link] Or copy and paste the URL below into your internet browser: [Direct URL] Please don't hesitate to contact me if you have questions or concerns. Thanks again! Jessica Wicks Director of Distanc e Learning, Multimedia, & Instructional Design (DLmid) Mid Michigan Community College 989.317.4601 or x101 jwicks@midmich.edu or wicksjes@msu.edu Follow the link below if you wish to opt out of this research project: [Link to unsubscribe] 146 APPENDIX D Notice of Group Assignment - Experimental Hello, [Name]. Thank you for consenting to be involved in research on the MMCC adjunct experience and the new Moodle Socialwall format plug -in. (Though I previously referred to this new Moodle format as the eCommunity format, representatives from Moodle HQ suggested the name change and it has recently been adopted by the developers.) Group Assignment You have been randomly assigned to Group 2, surveys plus Moodle Socialwall pilot. I know some participants requested a specific group, which is unfortunately not an option. A randomized design allows me to capture comprehensive feedback from a wider audience of users that more accurately reflects the larger stakeholder group. Your continued participation is, of course, completely optional. What You're Being Asked to Do I'm asking you to complete a survey this week and a follow -up survey in one month. Both are brief and I offer $5 each for their submission. Also, I'm also asking you to participate in a Mood le Socialwall pilot for one month. Me and about 50 of your MMCC adjunct peers will also participate. Topics may include the Socialwall format, the adjunct experience at MMCC, and other topics of group interest. To show that I value your time, I offer in termittent drawing s of 4 -$100 gift cards to those participating throughout the month. How extensively must you participate? It's up to you. This is not a course experience, but a social media experience. Some Facebook users, for example, login every da y and actively post. Others login every couple of days, passively reviewing recent conversations and occassionally posting. Find a model that remains engaged but balances well with your other life commitments. Next Steps (Action Necessary By Saturday, M arch 7 at midnight EST) Please take the following steps to signify continued involvement in this research: Step 1 - Complete the survey at the link below by Saturday, March 7 at midnight EST Survey Link: [Survey Link] Or copy and paste the URL below into your internet browser: [Direct URL] 147 Step 2 - A "Socialwall Testing Link" will appear for you near the top right of your Moodle space sometime on Monday morning. You must be logged into Moodle to see it. After you com plete the survey in step 1, please click on this link to enter our test installation of Moodle. Access the Adjunct Moodle Socialwall Pilot course in that test Moodle space by Saturday, March 7 at midnight EST. Please don't hesitate to contact me if you h ave questions or concerns. Thanks again! Jessica Wicks Director of Distance Learning, Multimedia, & Instructional Design (DLmid) Mid Michigan Community College 989.317.4601 or x101 jwicks@midmich.edu or wicksjes@msu.edu Follow the link below if you wish to opt out of this research project: [Link to unsubscribe] 148 APPENDIX E Pre -test Thank you in advance for responding to this brief survey. Please remember that participation is completely voluntary. You may opt out at any point and your choice to opt out will in no way impact your current or future opportunities at MMCC. Though this research is not anonymous, the researcher holds the data collected in confidence and will report results in a way that protects individual identities. Section 1/4 - MMCC Socialization Experience First, I'll ask about your socialization as an adjunct faculty member at MMCC. By socialization, I mean your sense of integration into the institution. I am aware that some of you hold multiple roles at MMCC. That will be considered in the analysis of any data collected. When considering ONLY your ongoing socialization experience as an adjunct faculty member at MMCC, to what extent do you agree or disagree with the following statements? Strongly Agree Agree Neither Agree nor Disagree Disagree Strongly Disagree If I were participating in a job search, my primary department would be one of my top 5 choices when considering places to work. ! ! ! ! ! I feel valued in my work environment. ! ! ! ! ! I feel an ownership of my primary department. Ownership, in this sense, means to feel like an integral contributor to a community. ! ! ! ! ! My loyalty to my primary department is high. Loyalty, in this sense, means having a sense of devoted attachment. ! ! ! ! ! I have been given adequate information on how to serve the institution effectively. ! ! ! ! ! I have been provided with the resources necessary to serve the institution effectively. ! ! ! ! ! The service expectations of my primary department have been clearly explained to me. ! ! ! ! ! I thoroughly understand the expectations required for continued employment as an adjunct. ! ! ! ! ! I feel connected to others in my work environment. ! ! ! ! ! I have frequent opportunities to socialize ! ! ! ! ! 149 with colleagues from my work environment. Socialize, in this sense, means to interact in a friendly manner. Opportunities to formally interact with colleagues during work hours are readily available to me. Formally, in this sense, means to interact on work -related topics, projects, committees, etc. ! ! ! ! ! I consider my MMCC colleagues to be friends as well as colleagues. ! ! ! ! ! Here, I want to explore themes from a previous survey on MMCC adjunct faculty socialization. When considering ONLY your ongoing socialization experience as an adjunct faculty member at MMCC, to what extent do you agree or disagree with the following statements? Strongly Agree Agree Neither Agree nor Disagree Disagree Strongly Disagree The students are the main reason that I teach. ! ! ! ! ! I receive adequate feedback from my instructional administrator/dean on my practice. ! ! ! ! ! I receive adequate feedback from full -time faculty in my department on my practice. ! ! ! ! ! Adjunct input is valued. ! ! ! ! ! Adjunct input helps shape professional development opportunities. ! ! ! ! ! I would participate in daytime opportunities for involvement. ! ! ! ! ! I would participate in evening opportunities for involvement. ! ! ! ! ! I would participate in online synchronous opportunities for involvement. Synchronous means in real time, such as web conferencing. ! ! ! ! ! I would participate in online asynchronous opportunities for involvement. Asynchronous means outside of real time, such as discussion forums. ! ! ! ! ! As an adjunct, I have no interest in developing strong social relationships with my MMCC colleagues. ! ! ! ! ! 150 Are there other aspects of your socialization as an adjunct faculty member at MMCC that you feel are important to your sense of integration into the institution? If so, note them below. If not, you may skip this question. Section 2/4 - Colleague Interactions Now, consider your interactions with colleagues at MMCC. By colleagues, I mean any members of our MMCC employee groups (adjunct or full -time faculty, hourly support staff, or administration). In a typical month, how frequently do you interact with MMCC colleagues in the following ways? Daily Weekly Monthly Less Than Monthly Face -to-Face ! ! ! ! Telephone ! ! ! ! Text Message or SMS ! ! ! ! Email ! ! ! ! Social Media (e.g. Facebook, Twitter, LinkedIn, etc.) ! ! ! ! When considering how typical interactions with MMCC colleagues may impact you as an adjunct faculty member, to what extent do you agree or disagree with the following statements? Strongly Agree Agree Neither Agree nor Disagree Disagree Strongly Disagree These interactions enhance my sense of ownership for my primary academic department. ! ! ! ! ! These interactions expose me to information that is relevant to my service as an adjunct. ! ! ! ! ! These interactions support my sense of connectedness to my MMCC colleagues. ! ! ! ! ! These interactions result in collaboration with colleagues to achieve shared goals. ! ! ! ! ! The content of these interactions directly applies to my daily experiences as an adjunct. ! ! ! ! ! These interactions connect my learning at MMCC with my learning outside of MMCC. ! ! ! ! ! The content of these interactions is customized to my needs as an adjunct. ! ! ! ! ! Are there other aspects of your interactions with colleagues at MMCC that you feel are important to your social experience at work? If so, note them below. If not, you may skip this question. 151 Section 3/4 - Social Media Use Lastly, I would like to gather baseline information about your social media use. In a typical week, select the types of social media that you use. (Check all that apply.) ! Blogging or Microblogging Platforms (e.g. Blogger, Twitter, Tumblr) ! Collaborative Editing (e.g. Googl e Drive, EtherPad) ! Photo or Video Sharing (e.g. YouTube, Instagram, Snapchat) ! Social Bookmarking (e.g. Pinterest, Delicious, Reddit) ! Social Networking Sites (e.g. Facebook, Google, LinkedIn) ! Other Social Media Used Weekly (Fill In.) ____________________ Are there other aspects of your social media use that you feel are important in planning adjunct support initiatives? If so, note th em below. If not, you may skip this question. Section 4/4 - Demographics Lastly, I'd like to collect basic demographic information. Data collected will be held in confidence and results will be reported in a way that protects individual identities. What is your age? (Whole numbers only.) What is your gender? ! Male ! Female ! I do not define myself by traditional societal gender categories. In which employee groups do you serve at MMCC? (Check all that apply.) ! Adjunct Faculty ! Administration ! Full -Time Faculty ! Hourly Support Staff To which group were you assigned for this research? ! Group 1 - Surveys Only ! Group 2 - Surveys Plus Moodle Socialwall Pilot Enter the 3 letter designator for the primary department at MMCC in which you teach. (e.g., ART, POL, IND) If you teach equal amounts in two or more departments, select the content area with which you feel most connected. In your current term of service with MMCC, for how many consecutive years have you held adjunct faculty member status? (Numbers only.) In terms of our active adjunct facult y pool, a "term of service" would end if the adjunct faculty person had not been assigned to a course for a full year. Are you currently teaching with MMCC this Winter semester? ! Yes ! No Are you currently teaching ONLY online with MMCC this Winter semester? ! Yes ! No Are you currently teaching ONLY off -campus with MMCC this Winter semester? ! Yes 152 ! No Are there other aspects of your demographics as an adjunct faculty member at MMCC that you feel are important in planning adjunct support init iatives? If so, note them below. If not, you may skip this question. 153 APPENDIX F Pre -test Reminder - Control Hello, [Name]. Thanks again for your consent to participate in the adjunct pilot of the new Moodle Socialwall format plug -in. I wrote you last week to ask you to tell you that you were assigned to Group 1 - the surveys only - and to ask you to complete the survey below. Just in case that email was lost in the shuffle, this is a reminder that you only have just over 24 hours to opt in at the link below before you're omitted from the included participants list. Please do the following by Saturday, March 7 at midnight EST to initiate participation in this research: Complete Survey 1 at the link below: Survey Link: [Survey Link] Or copy and paste the URL below into your internet browser: [Direct URL] Please don't hesitate to contact me if you have questions or concerns. Thanks again! Jessica Wicks Director of Distance Learning, Multimedia, & Instructional D esign (DLmid) Mid Michigan Community College 989.317.4601 or x101 jwicks@midmich.edu or wicksjes@msu.edu Follow the link below if you wish to opt out of this research project: [Link to unsubscribe] 154 APPENDIX G Pre -test Reminder - Experimental Hello, [Name]. Thanks again for your consent to participate in the adjunct pilot of the new Moodle Socialwall format plug -in. I wrote you last week to ask you to tell you that you were assigned to Group 2 - the surveys plus the Moodle pilot - and to ask you to take the two steps below. Just in case that email was lost in the shuffle, this is a reminder that you only have just over 24 hours to opt in at the link below before you're omitted from the included participants list. Though I'm ask ing you to participate across the course of a month by interacting with your peers in the Moodle social space, please remember that I'll be raffling off $100 gift cards to show appreciation for that participation. Please take the following steps by Saturd ay, March 7 at midnight EST to initiate participation in this research: Step 1: Complete Survey 1 at the link below: Survey Link: [Survey Link] Or copy and paste the URL below into your internet browser: [Direct URL] Step 2: Visit the Moodle Socialwall Adjunct Pilot space and get started. You can find this space by logging into Moodle and looking in the top right for the red text link that directs you to this space. Please don't hesitate to contact me if you have questions or concerns. Thanks again! Jessica Wicks Director of Distance Learning, Multimedia, & Instructional Design (DLmid) Mid Michigan Community College 989.317.4601 or x101 jwicks@midmich.edu or wicksjes@msu.edu Follow the link below if you wish to opt out of this research project: [Lin k to unsubscribe] 155 APPENDIX H Post -test Email Hello, [Name]. Thank you for your continued participation in this research on the MMCC adjunct experience and the new Moodle Socialwall format plug -in. The one -month research timeframe has now passed. To conclude this research, I am asking both groups to complete the final survey below at your earliest convenience: Action Necessary By Saturday, April 4 at midnight EST Complete the sur vey at the link below by Saturday, April 4: Survey Link: [Survey Link] Or copy and paste the URL below into your internet browser: [Direct URL] Also, as a final note, if you were part of the group that did surveys only and you'd now like a chance to tes t out this new Moodle tool, just send me an email request for access to a public demo space. Thanks again! Jessica Wicks Director of Distance Learning, Multimedia, & Instructional Design (DLmid) Mid Michigan Community College 989.317.4601 or x101 jwicks@midmich.edu or wicksjes@msu.edu Follow the link below if you wish to opt out of this research project: [Link to unsubscribe] 156 APPENDIX I Post -test Thank you in advance for responding to this brief survey. Please remember that participation is completely voluntary. You may opt out at any point and your choice to opt out will in no way impact your current or future opportunities at MMCC. Though this research is not anonymous, the researcher holds the data collected in confidence and will rep ort results in a way that protects individual identities. Section 1/4 - MMCC Socialization Experience First, I'll ask again about your socialization as an adjunct faculty member at MMCC. By socialization, I mean your sense of integration into the institution. I am aware that some of you hold multiple roles at MMCC. That will be considered in the analysis of any data collected. When considering ONLY your ongoing socialization experience as an adjunct faculty member at MMCC, to what extent do you agree or disagree with the following statements? Strongly Agree Agree Neither Agree nor Disagree Disagree Strongly Disagree I have frequent opportunities to socialize with colleagues from my work environment. Socialize, in this sense, means to interact in a friendly manner. ! ! ! ! ! If I were participating in a job search, my primary department would be one of my top 5 choices when considering places to work. ! ! ! ! ! I have been provided with the resources necessary to serve the institution effectively. ! ! ! ! ! I consider my work colleagues to be friends as well as colleagues. ! ! ! ! ! I thoroughly understand the expectations required for continued employment as an adjunct. ! ! ! ! ! I feel an ownership of my primary department. Ownership, in this sense, means to feel like an integral contributor to a community. ! ! ! ! ! I feel connected to others in my work environment. ! ! ! ! ! I have been given adequate information on how to serve the institution effectively. ! ! ! ! ! The service expectations of my primary department have been clearly explained to me. ! ! ! ! ! 157 I feel valued in my work environment. ! ! ! ! ! Opportunities to formally interact with colleagues during work hours are readily available to me. Formally, in this sense, means to interact on work -related topics, projects, committees, etc. ! ! ! ! ! My loyalty to my primary department is high. Loyalty, in this sense, means having a sense of devoted attachment. ! ! ! ! ! Section 2/4 - Colleague Interactions Now, consider your interactions with colleagues at MMCC. By colleagues, I mean any members of our MMCC employee groups (adjunct or full -time faculty, hourly support staff, or administration). In THIS PAST MONTH, how frequently did you interact with MMCC colleagues in the following ways? Daily Weekly Monthly Less Than Monthly Face -to-Face ! ! ! ! Telephone ! ! ! ! Text Message or SMS ! ! ! ! Email ! ! ! ! Social Media (e.g. Facebook, Twitter, LinkedIn, etc.) ! ! ! ! When considering how typical interactions in THIS PAST MONTH with MMCC colleagues may impact you as an adjunct faculty member, to what extent do you agree or disagree with the following statements? Strongly Agree Agree Neither Agree nor Disagree Disagree Strongly Disagree These interactions enhance my sense of ownership for my primary academic department. ! ! ! ! ! These interactions expose me to information that is relevant to my service as an adjunct. ! ! ! ! ! These interactions support my sense of connectedness to my MMCC colleagues. ! ! ! ! ! These interactions result in collaboration with colleagues to achieve shared goals. ! ! ! ! ! The content of these interactions directly applies to my daily experiences as an adjunct. ! ! ! ! ! These interactions connect my learning at MMCC with my learning outside of MMCC. ! ! ! ! ! 158 The content of these interactions is customized to my needs as an adjunct. ! ! ! ! ! Section 3/4 - Social Media Use Now, I would like to gather information about your recent social media use. In a typical week of THIS PAST MONTH, select the types of social media that you used. (Check all that apply.) ! Blogging or Microblogging Platforms (e.g. Blogger, Twitter, Tumblr) ! Collaborative Editing (e.g. Google Drive, EtherPad) ! Photo or Video S haring (e.g. YouTube, Instagram, Snapchat) ! Social Bookmarking (e.g. Pinterest, Delicious, Reddit) ! Social Networking Sites (e.g. Facebook, Google, LinkedIn) ! Other Social Media Used Weekly (Fill In.) ____________________ Remind me, to which group were you assigned in this study? ! Group 1 - Surveys Only ! Group 2 - Surveys Plus Moodle Socialwall Pilot Section 4/4 - Moodle Socialwall Finally, I would like to ask you some questions about your participation in the Moodle Socialwall format pilot. [These questions only required for experimental group participants.] What do you feel you gained from the Socialwall experience, if anything? When considering potential use of the Socialwall at MMCC, to what extent do you agree or disagree with the following s tatements? Strongly Agree Agree Neither Agree nor Disagree Disagree Strongly Disagree I look forward to future opportunities to use the Socialwall. ! ! ! ! ! The Socialwall could effectively support online instruction. ! ! ! ! ! The Socialwall could effectively support face -to-face instruction. ! ! ! ! ! The Socialwall could effectively support committee work. ! ! ! ! ! The Socialwall could effectively support student organizations. ! ! ! ! ! The Socialwall could effectively support colleague connectedness. ! ! ! ! ! The Socialwall could effectively support student connectedness. ! ! ! ! ! 159 The Socialwall is useful as a backchannel. Backchannel means a conversation that goes on alongside the primary activity, presentation, or discussion. ! ! ! ! ! When considering your participation in the Socialwall as it relates to your adjunct faculty role at MMCC, to what extent do you agree or disagree with the following statements? Strongly Agree Agree Neither Agree nor Disagree Disagree Strongly Disagree Participation enhanced my sense of ownership for my primary academic department. ! ! ! ! ! Participation exposed me to information that is relevant to my service as an adjunct. ! ! ! ! ! Participation supported my sense of connectedness with my MMCC colleagues. ! ! ! ! ! What features of the Socialwall experience did you feel were most important in supporting your sense of ownership for your primary academic department, if any? Please describe. What features of the Socialwall experience did you feel were most important in supporting your access to inf ormation relevant to your service as an adjunct faculty member, if any? Please describe. What features of the Socialwall experience did you feel were most important in supporting your sense of connectedness for your colleagues at MMCC, if any? Please de scribe. When considering your participation in the Socialwall as it relates to your adjunct faculty role at MMCC, to what extent do you agree or disagree with the following statements? Strongly Agree Agree Neither Agree nor Disagree Disagree Strongly Disagree Interactions resulted in collaboration with peers to achieve shared goals. ! ! ! ! ! The content of interactions directly applied to my daily experiences as an adjunct. ! ! ! ! ! Interactions connected my learning at MMCC with my learning outside of MMCC. ! ! ! ! ! The content of interactions was customized to my needs as an adjunct. ! ! ! ! ! What features of the Socialwall experience did you feel most supported collaboration with peers to achieve shared goals, if any? Please describe. 160 What features of the Socialwall experience did you feel most applied to your daily experiences as an adjunct, if any? Please describe. What features of the Socialwall experience did you feel most connected your learning at MMCC with your learning outsi de of MMCC, if any? Please describe. What features of the Socialwall experience did you feel most allowed you to customized interactions to your needs as an adjunct, if any? Please describe. If there were specific things you believe we should change about the Socialwall format in terms of the software, please discuss these things below. 161 APPENDIX J Post -test Reminder Hello, [Name]. This is just a quick reminder to complete the last survey in the Adjunct/Moodle research project by Saturday at midnight. Thank you for your continued participation in this research and thank you in advance for completing the final survey. In case my previous email was lost in the shuffle, a new link to that survey is below. Please take the survey below by Saturday, April 4 at midnight EST: Survey Link: [Survey Link] Or copy and paste the URL below into your internet browser: [Direct URL] Please don't hesitate to contact me if you have questions or co ncerns. Thanks again! Jessica Wicks Director of Distance Learning, Multimedia, & Instructional Design (DLmid) Mid Michigan Community College 989.317.4601 or x101 jwicks@midmich.edu or wicksjes@msu.edu Follow the link below if you wish to opt out of this research project: [Link to unsubscribe] 162 APPENDIX K Interaction Prompts Prompt #1 Your continued service is key in meeting student demand. Does anyone want to guess how many adjuncts are in our active pool at MMCC? What about the current adjunct to full -time ratio? What impacts, if any, do you think this staffing model has on your teaching, on your course design, or on general strategic directions in which our academic departments are headed? Prompt #2 I've started a Moodle Choice Poll below. If you opt into this poll by selecting a response by this Thursday, March 12 at 12PM EST, you will be entered into a drawing for a $100 gift card. Out of t hose who respond by that date, a winner will be selected through the aid of a random number generator. Poll: Do you intend to participate in our faculty professional development day on Wednesday, April 1, 2015? For the first time, four different break -out sessions for adjunct faculty will be available in the afternoon while full -time faculty attend academic senate. For those of you that haven't previously attended a PD Day, there is compensation available for your participation. A full schedule for this event will be sent out by Academic Council in the near future. Prompt #3 Your input on this Moodle Socialwall format will be critical in evaluating its function and potential use. I started a Google Doc, linked below, where I'd like to ask if you'd note some initial impressions as a group. Feel free to edit my starters as necessary. [Link to Google Doc.] Prompt #4 The second $100 gift card for your participation in this study will be drawn this week. Participate in the survey below by Friday at noon EST if you'd li ke to be included in that drawing . Q1: Some adjuncts express interest in being more included in Co llege activities, having more frequent opportunities to interact and contribute to organizational efforts. Do you share that interest? If so, what would that look like for you? Q2: We've discussed using online spaces like this SocialWall or tools like Goo gle Hangouts to help people be involved from a distance. Would you be interested in trying a live Google Hangouts meeting with me next week? If so, name some days & times below. Prompt #5 It seems it's easier to follow conversations and participate with d epth in we move the ÒbigÓ conversations we want to have into forums. So, keep using "the Wall" for what you 163 want and check out the forums below to be a part of the discussion on some larger themes weÕve explored here: [Link to Instructional Technology Con versations Forum] [Link to Teaching Best Practices Conversations Forum] [Link to Adjunct Support Conversations Forum] Prompt #6 The third $100 gift card for your participation in this study will be drawn this week. Participate in the glossary activity above by Friday at noon EST if you'd like to be included in that drawing . One of the neat things about social media is that it allows us to bring our formal learning together (work, school, etc.) with our i nformal learning (hobbies, interests, etc.). Have you ever used the Moodle glossary activity before? Let's try this as a way to bring these two spheres together and get to know more about one another. Create two glossary entries about your interests and you will be entered i nto this week's gift card drawing . See my entries in the glossary for inspiration on what to do and to learn a bit more about me. Prompt #7 Bauer & Erdogan (2014) describe socialization as "...the extent to which newcomers acquire the knowledge, skills, and functional understanding of their new jobs; make connections with others in the organization, and garner insight into the culture, processes, and people in their new organization." The focus in the literature is almost always on "new" employees. What do you think the average years of service is for our adjuncts at MMCC? If you are "new" to MMCC, how is that socialization occurring for you. If you are "not new" to MMCC, what are your thoughts about more focused socialization effor ts for those with enduring service? What would those look like? Prompt #8 The final $100 gift card for your participation in this study will be drawn this week. Participate in the forum linked above by Tuesday, March 31 at midnight EST to be included in that drawing . What should we do with this space and the Moodle Socialwall tool now? The study itself is over this week. I'll send out the final survey and then I'll compile notes on the ideas shared in this space for my research and to share recommendations with the College. I've shared some options below, certainly not the only options, so let me know what you think we should do by posting a reply in the forum linked above: ¥ We could delete this shell and not make further plans at this point. ¥ We could move this shell into regular Moodle and leave it open to all of you. (Would anyone actually even still use it without the prompts and incentives?) ¥ We could move this shell into regular Moodle and open it to even more people (like all adjuncts, all faculty, all college faculty -staff -admins, or whatever). ¥ We could start a ne w space with this same format in regular Moodle for specific audiences (like a department, all adjuncts, all faculty, or all faculty -staff -admins). ¥ We could do any number of other things that you name. 164 APPENDIX L Coding Schemas RQ 2: Initial Socialwall Interaction Codes Adjunct Manual Adjunct Campus Space Adjunct Events Adjunct Website Administration Application Appreciation for Peers Campus Activities Collaboration Communication Connection Course Load Discussion Dual Roles at MMCC Emotional Support Face -to-Face Formal -Informal Links Full -Time Faculty Hobbies/Interests Incentives Information Initiative Intimacy Involvement Learning Life Conflicts LMS Moodle Multiple Affiliations Onboarding Online Orientation Ownership Personaliz ation Processes Product Quality Professional Development Student Success Support Staff Technical Support Technology RQ 2: Emerging Socialwall Interaction Themes Personal C haracteristics & Initiative Codes: Hobbies & Interests, Initiative, Learning, Ownership ¥ To Be Active in Connecting ¥ To Be Active in Pursuing Necessary Information & Resources ¥ To Own OneÕs Role in the Institution/Department Information Codes: Adjunct Manual, Adjunct Campus Space, Adjunct Events, Adju nct Website, Application, Campus Activities, Formal -Informal Links, Incentives, Information, Onboarding, Orientation, Personalization, Processes, Product Quality, Professional Development, Student Success ¥ To Foster Connection ¥ To Provide Necessary & Relevan t Information ¥ To Value Adjuncts Through Inclusion & Opportunities Conn ection with Colleagues Codes: Administration, Appreciation for Peers, Collaboration, Communication, Connection, Discussion, Emotional Support, Face -to-Face, Full -Time Faculty, Intimacy, Involvement, Support Staff 165 ¥ To Provide Necessary Information & Feedback ¥ To Foster Connection & Support ¥ To Contribute to Department/Institution ¥ To Value Adjuncts Through Inclusion & Opportunities Technology Codes: LMS, Moodle, Online, Technical Support, Technology ¥ To Provide Necessary Resources & Support ¥ To Value Adjuncts Through Technology -Mediated Inclusion & Opportunities Challenges to Onboarding Codes: Course Load, Dual Roles at MMCC, Life Conflicts, Multiple Affiliations ¥ To Acknowledge AdjunctsÕ Div erse Experience & Commitments ¥ To Value Adjuncts Through Inclusion & Opportunities RQ 3: Initial Socialwall Codes Regarding Socialwall Positivity Uncertainty Purpose/Utility Potential Functionality Excitement RQ 3: Emerging Socialwall Themes Regarding Socialwall ¥ Purpose (1 8) - Inquiries about Socialwall purpose/ goal or suggestions for its use along with uncertainty or lack of value. ¥ Positivity /Potential (31) - Perceived value and/or impact (present or future) from Socialwall experience or speculation on value of Socialwall across time . ¥ Functionality (11) - Discussion of Socialwall features, function, & effectiveness. RQ 3: Initial Posttest Codes Regarding Socialwall Accessibility Collaboration Connection Help Information Lack of Participation None Sharing Viewpoints Technology 166 RQ 2: Emerging Posttest Themes Regarding Socialwall Connection Codes: Collaboration, Connection, Help, Sharing Viewpoints ¥ Connection and sharing with peers. ¥ Support from peer s. ¥ Exposure to diverse perspectives. None Codes: None ¥ There were no noted gains. ¥ There were no important features. Information Codes: Accessibility, Information ¥ It was readily accessible. ¥ Relevant information was shared. Lack of Participation Codes: Lack of Participation ¥ Acknowledgement of failure to engage. Technology Codes: Technology ¥ Exposure to, and practice with, a new tool. ¥ Greater understanding of social media tools. 167 REFERENCES 168 REFERENCES Alexander, J. C., Jr. (1992). Mentoring: On the road to tenure and promotion. ACA Bulletin, 79 , 54-58. American Association of University Professors (AAUP). (2010). Tenure and teaching - intensive appointments. Retrieved from http://www.aaup.org/report/tenure -and -teaching -intensive -appointments. American Federation of Teachers (AFT). (2010). American Academic: A National Survey of Part -Time/Contingent Faculty (Vol. 2). Arnold, N. & Paulus, T. (2010). Using a social networking site for experiential learning: App ropriating, lurking, modeling and community building. Internet and Higher Education, 13 , 188-196. Ashforth, B. E., Sluss, D. M., & Saks, A. M. (2007). Socialization tactics, proactive behavior, and newcomer learning: Integrating socialization models. Journal of Vocational Behavior, 70 (3), 447-462. Baldwin, R. G. & Wawrzynski, M. R. (2011). Contingent faculty as teachers: What we know, what we need to know. American Behavioral Scientist, 55 (11), 1485-1509. Bandura, A. (1986). Social foundations of through and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice -Hall. Baron -Nixon, L. (2007). Connecting non full -time faculty to institutional mission: A guidebook for college/university administrators and faculty developers. Sterling, VA: Stylus. Bashir, M., Afzal, M. T., & Azeem, M. (2008). Reliability and validity of qualitative and operational research paradigm. Pakistan Journal of Statistics and Operation Research, IV (I), 35-45. Bates, M. A. (2012). Understanding burnout and promoting engagement among adjunct faculty in community colleges. (Doctoral dissertation.) Retrieved from Digital Commons@NLU Bauer, T. N., Bodner, T., Erdogan, B., Truxillo, D. M., & Tucker, J. S. (2011). Newcomer adjustment during organizational socialization: A meta -analytic review of antecedents, outcomes, and methods. Journal of Applied Psychology, 92 (3), 707-721. Bauer, T. N. & Erdogan, B. (2014). Delineating and reviewing the role of newcomer 169 capital in organization al socialization. Annual Review of Organizational Psychology and Organizational Behavior, 1 , 439-457. Beach, R. & Doerr -Stevens, C. (2011). Using social networking for online role -plays to develop studentsÕ argumentative strategies. Journal of Educati onal Computing Research, 45 (2), 165-181. Beaudoin, M. F. (2002). Learning or lurking: Tracking the ÒinvisibleÓ online student. Internet and Higher Education, 5 , 147-155. Birnie, S. A. & Horvath, P. (2002). Psychological predictors of internet social communication. Journal of Computer -Mediated Communication, 7 (4). Brady, K. P., Holcomb, L. B., & Smith, B. V. (2010). The use of alternative social networking sites in higher educational settings: A case study of the e -learning benefits of Ning in edu cation. Journal of Interactive Online Learning, 9 (2), 151-170. Burke, M., Marlow, C., & Lento, T. (2010). Social network activity and social well - being. Proceedings from CHI 2010: SIGCHI Conference on Human Factors in Computing Systems . Atlanta, GA. Burnstad, H. M. (2002). Part -time faculty development at Johnson County Community College. New Directions for Community Colleges, 120 , 17-26. Byrne, D. N. (2007). Public discourse, community concerns, and civic engagement: Exploring black social networking traditions on BlackPlanet.com. Journal of Computer -Mediated Communication, 13 (1), 319-340. Cawyer, C. S. & Friedrich, G. W. (1998). Organizational socialization: Processes for new communication faculty. Communication Education, 47, 234 Ð245. Cawyer, C. S., Simonds, C., & Davis, S. (2010). Mentoring to facilitate socialization: The case of the new faculty member. International Journal of Qualitative Studies in Education, 15 (2), 225-242. Charlier, H. & Williams, M. (2011). The reliance on and demand for contingent faculty membersin AmericaÕs rural, suburban, and urban community colleges. Community College Review, 39 (2), 160-180. Church, A. H. (1993). Estimating the effect of incentives on mail survey response rates: A meta -analysis. Public Opinion Quarterly, 57 , 62-79. Cooper -Thomas, H., Anderson, T. N., & Cash, M. (2011). Investigating organizational socialization: A fresh look at newcomer adjustment strategies. Personnel Review, 41 (1), 41-55. 170 Creswell, J. W. (2014). Research design: Qualitative, quantitative, and mixed methods approaches (4th ed.). Thousand Oaks, CA: Sage Publications, Inc. Crowdsourcing. (n.d.). In Wikipedia. Retrieved July 22, 2014, from http://en.wikipedia.org/wiki/Crowdsourcing Davis, C. H. F., Deil -Amen, R., Rios -Aguilar, C., & Gonzalez Canche, M. S. (2012). Social media in higher education: A literature review and research directions. Retrieved from http://works.bepress.com/hfdavis/2/ Dillman, D. A., Smyth, J. D., & Christian , L. M. (2009). Internet, Mail, and Mixed -Mode Surveys: The Tailored Design Method. Hoboken, NJ: Wiley and Sons. Dub”, L., Bourhis, A., & Jacob, R. (2005). The impact of structuring characteristics on the launching of virtual communities of practice. Journal of Organizational Change Management, 18 (2), 145Ð166. Eagan, M. K. & Jaeger, A. J. (2008). Closing the gate: Part -time faculty instruction in gatekeeper courses and first -year persistence. New Directions for Teaching and Learning, 115 , 39-53. Ellison, N. B., Steinfield, C., & Lampe., C. (2007). The benefits of Facebook Òfriends:Ó Social capital and college studentsÕ use of online social network sites. Journal of Computer -Mediated Communication, 12 (4), 1143-1168. Ellison, Wohn, & Greenhow ( 2014). AdolescentsÕ visions of their future careers, educational plans, and life pathways: The role of bridging and bonding social capital experiences. Journal of Social and Personal Relationships, 31 (4), 516-534. Fang, R., Duffy, M. K., & Shaw, J. D. (2011). The organizational socialization process: Review and development of a social capital model. Journal of Management, 37 (1), 127-152. Finucane, M. O. & Algren, M. E. (1997). A pilot study investigation of the socialization of part -time faculty members in academic settings. Paper presented at the 83rd Annual Meeting of the National Communication Association . Chicago, IL. Gao, F., Luo, T., & Zhang, K. (2012). Tweeting for learning: A critical analysis of research on microblogging in education published in 2008 -2011. British Journal of Educational Technology, 43 (5), 783-801. Gappa, J. M., Austin, A. E., & Trice, A. G. (2007). Rethinking Faculty Work . San Francisco: Wiley and Sons. 171 Gillespie, K. J. & Robertson, D. L. (Eds.). (2010). A guide to faculty development. San Francisco, CA: John Wiley and Sons. Glesne, C. (2011). Becoming qualitative researchers: An introduction (4th ed.). Boston, MA: Pearson Education. Gonzales, J. (2009). Connecting with part -timers is key cha llenge for community colleges, survey finds. The Chronicle of Higher Education. Retrieved from http://chronicle.com/article/Connecting -With-Part -Timers -Is/49139/ Grasmuch, S., Martin, J., & Zhao, S. (2009). Ethno -racial identity displays on Facebook. Journal of Computer -Mediated Communication, 15 (1). Greenhow, C. (2010). Youth as content producers in a niche social network site. New Directions for Youth Development, 128 , 55-63. Greenhow, C. & Burton, L. (2011). Help from my ÒFriends:Ó Social ca pital in the social Network sites of low -income high school students. Journal of Educational Computing Research, 45 (2), 223-245. Greenhow, C. & Robelia, B. (2009). Old communication, new literacies: Social network sites as social learning resources. Journal of Computer -Mediated Communication, 14, 1130-1161. Gruzd, A., Staves, K., & Wilk, A. (2012). Connected scholars: Examining the role of social media in research practices of faculty using the UTAUT model. Computers in Human Behavior, 28 (6), 2340-2350. Heiberger, G. & Harper, R. (2008). Have you Facebooked Astin lately? Using technology to increase student involvement. New Directions for Student Services, 124, 19-35. Hessler, K. & Ritchie, H. (2006). Recruitment and retention of novice f aculty. The Journal of Nursing Education, 45 (5), 150-154. Higher Education Research Institute (HERI). (2007). College freshmen and online social networking sites. Retrieved from http://www.heri.ucla.edu/PDFs/pubs/briefs/brief -091107-SocialNetworking .pdf Hummel, H. G. K., Burgos, D., Tattersall, C., Brouns, F., Kurvers, H., & Koper, R. (2005). Encouraging contributions in learning networks using incentive mechanisms. Journal of Computer Assisted Learning, 21 , 355-365. Ito, M., Hurst, H., Bittanti , M., boyd, d., Herr -Stephenson, B., Lange, P. G., . . . Tripp, L. (2008). Living and learning with new media: Summary of findings from the digital youth project. White paper, the John D. and Catherine T. MacARthur 172 Foundation Reports on Digital Media an d Learning. Retrieved from http://files.eric.ed.gov/fulltext/ED536072.pdf. Jacoby, D. (2006). Effects of part -time faculty employment on community college graduation rates. The Journal of Higher Education, 77 (6), 1081-1103. Jolley , M. R., Cross, E., & Bryant, M. (2013). A critical challenge: The engagement and assessment of contingent, part -time adjunct faculty professors in United States community colleges. Community College Journal of Research and Practice, 38 , 218-230. Junco , R. (2012). The relationship between frequency of Facebook use, participation in Facebook activities, and student engagement. Computers & Education, 58 (1), 162-171. Junco, R., Heibergert, G., & Loken, E. (2011). The effect of Twitter on college stud ent engagement and grades. Journal of Computer Assisted Learning, 27 (2), 119-132. Kammeyer -Mueller, J. D., Livingston, B. A. & Liao, H. (2005). Perceived similarity, proactive adjustment, and organizational socialization. Journal of Vocational Behavi or, 78 (2), 225-236. Kaplan, A. M. & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities of social media. Business Horizons, 53 (1), 61. Kay, R. (2007). QuickStudy: Online social networks. These sites can facilitate connections in your industry or around the world. Computerworld. Retrieved from http://www.computerworld.com/article/2552185/networking/online -social -networks.html Kezar, A. (2013). Examining non -tenure track faculty perceptions of how departmental policies and practices shape their performance and ability to create student learning at four -year institutions. Research in Higher Education, 45 (5), 571-598. Kirschner, P. A. & Karpinski, A. C. (201 0). Facebook¨ and academic performance. Computers in H uman Behavior, 26 (6), 1237-1245. Kramer, M. W., Callister, R. R., & Turban, D. B. (1995). Information -receiving and information -giving during job transitions. Western Journal of Communication, 59 (2), 151-170. Kramer, A. L., Gloeckner, G. W., & Jacoby, D. (2014). Roads scholars: Part -time faculty job satisfaction in community colleges. Community College Journal of Research and Practice, 38 (4), 287-299. Kuh, G. D. (2009). The national survey of student engagement: Conceptual and 173 empirical f oundations. In R. Gonyea and G. Kuh (Eds.), Using student engagement in institutional research. New Directions for Institutional Research , 141, (pp. 5-20). San Francisco, CA: Jossey -Bass. Lally, P., van Jaarsveld, C. H. M., Potts, H. W. W., & Wardle, J . (2009). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40 (6), 998-1009. Landrum, R. E. (2009). Are there instructional differences between full -time and part - time faculty? College Teaching, 57 (1), 23-26. Langen, J. M. (2011). Evaluation of adju nct faculty in higher education institutions. Assessment & Evaluation in Higher Education, 36 (2), 185-196. Larcara, M. (2010). Forecasting online adjunct needs: A delphi study. Retrieved from ProQuest Digital Dissertations. (3426718) Lave, J. & Wegner, E. (1991). Situated Learning: Legitimate Peripheral Participation. Cambridge: Cambridge University Press. Leslie, D. W., & Gappa, J. M. (2002). Part -time faculty: Competent and committed. New Directions for Community Colleges, 118 , 59-68. Louis, M. R. (1980). Surprise and sense making: What newcomers experience in entering unfamiliar organizational settings. Administrative Science Quarterly, 25 , 226-251. McCoy, D. (2006). Entering the academy: Exploring the socialization experiences of African American male faculty (Doctoral dissertation). Retrieved from lsu.edu. McKenna, K. Y. A. & Bargh, J. A. (2000). Plan 9 from cyberspace: The implications of the i nternet for personality and social psychology. Personality and Social Psychology , 4 (1), 57-75. Meixner, C., Kruck, S. E., & Madden, L. T. (2010). Inclusion of part -time faculty for the benefit of faculty and students. College Teaching, 58 (4), 141-147. Minocha, S. (2009). Role of social software tools in education: A literature review. Education + Training, 51 , 353-369. Moodle Facebook/Social Course Format (2014). In Moodle.org. Retrieved June 29, 2014, from https://moodle.org/mod/forum/discuss.php?d=257998 Moran, M., Seaman, J., & Tinti -Kane, H. (2011). Teaching, Learning, and Sharing: How TodayÕs Higher Education Faculty Use Social Media. Research report published by Pearson, The Babson Survey Research Gr oup, and Converseon. 174 Mueller, B., Mandernach, B. J., & Sanderson, K. (2013). Adjunct versus full -time faculty: Comparison of student outcomes in the online classroom. MERLOT Journal of Online Learning and Teaching, 9 (3). Murray, J. P. (2001). Facul ty development in publicly supported 2 year colleges. Community College Journal of Research and Practice, 25 , 487-502. Nelson, D. L. & Quick, J. C. (1991). Social support and newcomer adjustment in organizations: Attachment theory at work? Journal of Organizational Behavior, 12 (6), 543-554. Nonnecke, B. & Preece, J. (2001). Why lurkers lurk. Paper presented at Americas Seventh Conference on Information Systems (AIS, 2001), Boston, MA. Ostroff, C. & Kozlowski, S. W. J. (1993). The role of mentoring in the information gathering processes of newcom ers during early organizational socialization. Journal of Vocational Behavior, 42 (2), 170-183. Pallant, J. (2013). SPSS: Survival manual (5th ed.). Ber kshire, England: McGraw -Hill Education. Paul, J. A., Baker, H. M., & Cochran, J. D. (2012). Effect of online social networking on student academic performance. Computers in Human Behavior, 28 (6), 2117-2127. Pempek, T. A., Yermolayeva, Y. A., & Calvert, S. L. (2009). College studentsÕ social networking experiences on Facebook. Journal of Applied Developmental Psychology, 30 (3), 227-238. Perez -Pena, R. (2013). College enrollment falls as economy recovers. New York Times. Retrieved August 1, 2013, from http://www.nytimes.com/2013/07/26/education/in -a-recovering -economy -a-decline -in-college -enrollment.html Perna, F. M., Lerner, B. M., & Yura, M. T. (1995). Mentoring and career development among university faculty. Journal of Education, 177 (2), 31-45. Pimmer, C., Linxen, S., & Grıhbiel, U. (2012). Facebook as a learning tool? A case study on the appropriation of social network sites from mobile phones in developing countries. British Journal of Educational Technology, 43 (5), 726-738. Plug-in (computing). (n.d.). In Wikipedia. Retrieved August 17, 2014, from http://en.wikipedia.org/wiki/Plug -in_(computing) Preece, J., Nonnecke, B., & Andrews, D. (2004). The top five reasons for lurking: 175 Improving community experiences for everyone. Computers in Human Behavior, 20, 01-223. Ranieri, M., Manca, S., & Fini, A. (2012). Why (and how) do teachers engage in social networks? An exploratory study of professional use of Facebook and its implications for lifelong learning. British Journal of Educational Technology, 43 (5), 754-769. Rau, P. P., Gao, Q., & Ding, Y. (2008). Relationship between the level of intimacy and lurking in online social network services. Computers in Human Behavi or, 24 , 2757-2770. Robelia, B. A., Greenhow, C., & Burton, L. (2011). Environmental learning in online social networks: Adopting environmentally responsible behaviors. Environmental Education Research, 17 (4), 553-575. Ronco, S. L. & Cahill, J. (2004). Does it matter whoÕs in the classroom? Effect of instructor type on student retention, achievement and satisfaction. Paper presented at the 44th Annual Forum of the Association for Institutional Research . Boston, MA. Rorhle, B. & Sommer, G. (1994). Soc ial support and social competences: Some theoretical and empirical contributions to their relationship. In Frank Nestmann and Hurrelmann, Klaus. (Eds.), Social Networks and Social Support in Childhood and Adolescence. Berlin, Germany: De Gruyter, 111 -130. Roueche, J. E., Roueche, S. D., & Milliron, M. D. (1996). Identifying the strangers: Exploring part -time faculty integration in American community colleges. Community College Review, 23 (4), 33-48. Rowlands, I., Nicholas, D., Russell, B., Canty , N., & Watkinson, A. (2011). Social media use in the research workflow. Learned Publishing, 24 (3), 183-195. Saks, A. M. (1994). Moderating effects of self -efficacy for the relationship between training method and anxiety and stress reactions of newcomers. Journal of Organizational Behavior, 15 , 639-654. Saks, A. M. & Ashforth, B. E. (1997). Organizational socialization: Making sense of the past and present as a prologue for the future. Journal of Vo cational Behavior, 51 , 234-279. Saks, A. M., Uggerslev, K. L., Fassina, N. E. (2007). Socialization tactics and newcomer adjustment: A meta -analytic review and test of a model. Journal of Vocational Behavior, 70 , 413-446. Schellens, T. & Valcke, M. (2 005). Collaborative learning in asynchronous discussion 176 groups: What about the impact on cognitive processing? Computers in Human Behavior, 21 (6), 957-976. Schrodt, P., Cawyer, C. S., & Sanders, R. (2003). An examination of academic mentoring behaviors and new faculty membersÕ satisfactio n with socialization and tenure and promotion processes. Communication Education, 52 (1), 17-29. Schroeder, J. & Greenbowe, T. J. (2009). The chemistry of Facebook: Using social networking to create an online community for the organic chemistry laboratory. Innovate, 5 (4). Scott, K. L. (2010). Managing communi ty college adjuncts in the 21 st century. Retrieved from ProQuest Digital Dissertations. (3427253) Shannon, D. A. (2007). The socialization of adjunct faculty into the academic culture of a public community college campus . Retrieved from ProQuest ETD Collection for FIU . (AAI3298597) Singer, E., Van Hoewyk, J., Gebler, N., Raghunathan, T., & McGonagle, K. (1999). The effect of incentives on response rates in interviewer -mediated surveys. Journal of Official Statistics, 15 (2), 217-230. Statistic Brain. (2014a). Facebook statistics. Retrieved July 21, 2014 from http://www.statisticbrain.com/facebook -statistics/ Statistic Brain. (2014b). Twitter statistics. Retrieved July 21, 2014 from http://www.statisticbrain.com/twitter -statistics/ Steinfield, C., Ellison, N. B., & Lampe, C. (2008). Social capital, self -esteem, and use of Online social network sites: A longitudi nal analysis. Journal of Applied Developmental Psychology, 29 , 434-445. Stout, R. A. (2013). Working at home: Exploring gender roles and burnout among online adjunct faculty at a proprietary postsecondary institution. Retrieved from ProQuest Digital Dissertations. (3577984) Subrahmanyan, K., Reich, S. M., Waechter, N., & Espinoza, G. (2008). Online and offline social networks: Use of social networking sites by emerging adults. Journal of Applied Developmental Psychology, 29 (6), 420-433. Tess, P. A. (2013). The role of social media in higher education classes (real and virtual) - A literature review. Computers in Human Behavior, 29 (5), A60 -A68. Thematic analysis. (n.d.). In Wikipedia. Retrieved June 8, 2015, from http://en.wikipedia.org/wiki/Thematic_analysis 177 Thyer, B. A., Myers, L. L., & Nugent, W. R. (2011). Do regular social work faculty earn better student course evaluations than do adjunct faculty or doctoral students? Journal of Teaching in Social Work, 31 (4), 365-377. Tierney, W. G. & Rhoads, R. A. (1994). Enhancing promotion, tenure and beyond: Faculty socialization as a cultural process. ASHE -ERIC Higher Education Report 6. Trowler, P. & Knight, P. (1999). Organizational socialization and inducti on in universities: Reconceptualizing theory and practice. Higher Education, 37 , 177-195. Umbach, P. D. (2007). How effective are they? Exploring the impact of contingent faculty on undergraduate education. Review of Higher Education, 30 (2), 91-118. Valenzuela, S., Park, N., & Kee, K. F. (2009). Is there social capital in a social network site?: Facebook use and college studentsÕ life satisfaction, trust, and participation. Journal of Computer -Mediated Communication, 14 (4), 875-901. Valkenburg, P . M., Peter, J., & Schouten, A. P. (2006). Friend networking sites and their relationship to adolescentsÕ well -being and social self -esteem. CyberPsychology & Behavior, 9 (5), 584-590. Van Maanen, J. & Schein, E. H. (1979). Toward a theory of organizational socialization. In B. M. Staw (Ed.), Research in Organizational Behavior , (pp. 209-264). Greenwich, CT: JAI Press. Velez, A. M. (2008). Evaluating research methods: Assumptions, strengths, and weaknesses of three educational research parad igms. Academic Exchange Extra, 9. Retrieved August 16, 2014 from http://www.unco.edu/ae -extra/2008/9/velez.html Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of information technology: Toward a unified view. Management Information Systems Quarterly, 27 (3), 425-478. Vygotsky, L. (1978). Mind in Society. London: Harvard University Press. Walther, J. B. & Parks, M. R. (2002). Cues filtered out, cues filtered in: Computer - mediated communication and relationships. In M. L. Knapp & J. A. Daly (Eds.), Handbook of Interpersonal Communication (3rd ed., pp. 529 -563). Thousand Oaks, CA: Sage. Wanous, J. P. (1993). Newcomer orientation programs that facilitate organizational 178 entry. in Schuler, H., Farr, J. L., & Smith, M. (Eds.) Personnel Selection and Assessment: Individual and Organizational Perspectives . Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Watts, G. E. & Hammons, J. O. (2002). Professional development: Setting the context. New Directions for Co mmunity Colleges, 120 , 5-10. Wegner, E. (2007). Communities of practice: A brief introduction. Retrieved from http://wenger -trayner.com/introduction -to-communities -of-practice/ Welch, G. F. (2002). A new faculty orientation program: Building a core of new faculty to shape the future of the college. New Directions for Community Colleges, 120, 11-16. West, M. M., Borden, C., Bermudez, M., Hanson -Zalot, M., Amorim, F., & Marmion , R. (2009). Enhancing the clinical adjunct role to benefit students. Journal of Continuing Education in Nursing, 40 (7), 305-310. West, A., Lewis, J., & Currie, P. (2009). StudentsÕ Facebook ÔfriendsÕ: Public and private spheres. Journal of Youth Studies, 12 (6), 615-627. Wicks, J. M. (2013). The socialization of contingent faculty members: The potential of social media. (Unpublished research practicum). Michigan State University, East Lansing, MI. Wright, K. B. (1999). Computer -mediated supp ort groups: An examination of relationships among social support, perceived stress, and coping strategies. Communication Quarterly, 47 (4), 402-414. Ziegler, C. A. & Reiff, M. (2004). Adjunct mentoring: A responsibility, not a choice. Paper presented at the American Association of Colleges for Teacher Education Annual Conference , Chicago, IL, 7 -10 February. Ziegler, C. A. & Reiff, M. (2005). Adjunct mentoring in SOEÕs: The whyÕs, how toÕs, and your 1st steps. Paper presented at the A merican Associ ation of Colleges for Teacher Education Annual Conference , Washington, DC, 19 -23, February. Ziegler, C. A. & Reiff, M. (2006). Adjunct mentoring, a vital responsibility in a changing educational climate: The Lesley University Adjunct Mentoring Program. Mentoring & Tutoring, 14 (2), 247-269. Zimmerman, K., & Struthers, S. A. (2011). Adjunct faculty certification 101: Introduction to teaching and learning. Retrieved from http://corescholar.libraries.wright.edu/art/19